Association between pulmonary function and sputum biomarkers in cystic fibrosis

RATIONALE

Sputum biomarkers of infection and inflammation are noninvasive measures that enable quantification of the complex pathophysiology of cystic fibrosis (CF) lung disease. Validation of these biomarkers as correlates of disease severity is a key step for their application.

OBJECTIVES

We constructed a large database from four multicenter studies to quantify the strength of association between expectorated sputum biomarkers and FEV(1.)

METHODS

FEV(1) (range, 25-120% predicted) and quantitative data on expectorated sputum biomarkers including free neutrophil elastase, IL-8, neutrophils, Pseudomonas aeruginosa, and Staphylococcus aureus were obtained from 269 participants (ages, 9-54 years) from 33 centers. Cross-sectional and longitudinal statistical analyses were performed to estimate associations between the markers and FEV(1), including the use of multivariable analyses.

RESULTS

Elastase was negatively correlated with FEV(1) (correlation [r] = -0.35; 95% confidence interval [CI]: -0.46, -0.22). On average, patients with CF who differed in their elastase measurements by 0.5 log differed in their FEV(1) values by -7.3% (95% CI: -9.7, -4.6). Neutrophil counts and IL-8 were also each negatively correlated. In a multivariable regression, elastase and neutrophil counts were able to explain the majority of variation in FEV(1). Elastase was further shown to have a significant longitudinal association with FEV(1), specifically a -2.9% decline in FEV(1) (95% CI: -5.0, -0.9) per 1-log increase in elastase. Although correlated with FEV(1), bacterial densities were unable to explain clinically meaningful differences in FEV(1) within and across patients.

CONCLUSIONS

These data support the role of sputum biomarkers as correlates of disease severity in a diverse CF population.

Laboratory parameter profiles among patients with cystic fibrosis

BACKGROUND

Clinical trials in cystic fibrosis (CF) currently use laboratory-specific reference ranges to evaluate chemistry and hematology measurements. Laboratory-specific normal reference ranges may not accurately reflect what is abnormal but clinically insignificant among CF patients.

METHODS

To address this concern, data from the Phase III trial of inhaled tobramycin in CF patients was used to describe the distribution and variability of laboratory parameters. The laboratory specimens were analyzed at a central laboratory after being obtained at baseline and throughout the 24-week trial.

RESULTS

At the time of entry into the clinical trial, 91% (463 of 508) of patients had at least a single value outside the normal range. Liver function tests (AST, ALT) were above the normal range in 16% and 12% of the patients respectively, with 2.4% of patients having an AST>2.0 times the upper limit of normal. Of the 243 patients on placebo, 242 (99.6%) had at least one laboratory parameter that changed from normal to abnormal during the 24-week follow-up period. Of those same placebo patients, 11.5% (N=28) had a laboratory parameter change from a Common Toxicity Criteria (CTC) grade 0 to grade 2 or higher during follow-up.

CONCLUSIONS

Patients with CF frequently have laboratory values outside the normal range and have significant longitudinal variability of laboratory values. Interpretation of adverse events in the clinical trial setting may be complicated by the underlying high rates of some laboratory abnormalities in the CF population. This data was presented in poster format at the American Thoracic Society International Conference, Atlanta, USA, 2002, appearing subsequently in the Conference proceedings [Goss CH, Mayer-Hamblett N, Yunker A, Waltz DA, Kronmal RA, Ramsey BW. Laboratory parameter profiles among patients with cystic fibrosis. Am J Rep Crit Care Med 2002;165(8):A283].

Heterogeneity of treatment response to azithromycin in patients with cystic fibrosis

RATIONALE

We recently reported a randomized, placebo-controlled trial of azithromycin in patients with cystic fibrosis (CF) that demonstrated a 6.2% improvement in the 168-d relative change in FEV1 among azithromycin participants compared with placebo participants.

OBJECTIVES

In the current analyses, heterogeneity of treatment response and the association between FEV1 and the risk of pulmonary exacerbations were investigated.

METHODS

The time to first pulmonary exacerbation, hospitalization rates, and antibiotic use were compared between participants categorized by their relative change in FEV1 % predicted (>or= 5 vs. < 5% improvement) at Day 168. Pulmonary function and exacerbation responses were compared in subgroups of participants characterized by long-term concomitant medications and baseline lung function.

MEASUREMENTS

All available data from the 185 randomized participants in the azithromycin trial were included in these analyses.

MAIN RESULTS

Compared with placebo participants, a reduced risk of pulmonary exacerbations was observed both among azithromycin participants with >or= 5% and those with < 5% relative improvement in FEV1. Similarly, decreased hospitalization rates and decreased use of oral quinolone and nonquinolone antibiotics were observed in azithromycin participants regardless of improvement in FEV1. Subgroup analyses demonstrated that overall, participants on long-term aerosolized tobramycin and/or rhDNase had worse baseline lung function, but still benefited from azithromycin, as evidenced by a lower risk of exacerbations.

CONCLUSIONS

Azithromycin participants experienced benefits in exacerbation parameters regardless of FEV1 response or subgroup. These data have implications for clinical practice and the design of clinical trials.

Randomized, double-blind, placebo-controlled, dose-escalating study of aerosolized interferon gamma-1b in patients with mild to moderate cystic fibrosis lung disease

Interferon gamma-1b (IFN-gamma1b) is a pleiotropic cytokine with immunomodulatory activities that could decrease bacterial burden, inflammation, and obstruction in patients with CF. Patients with CF (> or =12 years old, FEV1 > or =40% predicted) were randomly assigned to sequential dose cohorts inhaling 500 microg IFN-gamma1b, 1,000 microg IFN-gamma1b, or placebo by Respirgard II nebulizer thrice weekly for 12 weeks. Sputum bacterial density and spirometry were measured. Safety, antibiotic use, hospitalization, and sputum neutrophils, elastase, DNA, IL-8, and myeloperoxidase were also evaluated. Sixty-six patients (mean age, 24 years, with mean baseline FEV1 of 74 +/- 20 (SD) percent predicted) were studied. One patient had bronchospasm after the first dose of IFN-gamma1b; the overall withdrawal rate was 15% (5 in the placebo group, 2 in the 500-microg IFN-gamma1b group, and 3 in the 1,000 microg IFN-gamma1b group). The 500-microg IFN-gamma1b dose was well-tolerated, but the 1,000-mug dose cohort, who had a higher baseline bacterial density than placebo patients (mean difference, 1.2 log(10) CFU/g sputum, 95% confidence interval (CI), 0.1,2.8, P=0.04), had 24% more hospitalizations for exacerbation than placebo patients (95% CI, 2,45%, P=0.05). There was a 0.12-l difference between the 500-microg IFN-gamma1b and placebo groups with respect to the 12-week change in FEV1 (active group minus placebo group, 95% CI, -0.03,0.26, P=0.11), as compared to a 0.01-l difference between the 1,000-microg IFN-gamma1b and placebo groups (95% CI, -0.16,0.17, P=0.96). No effects of IFN-gamma1b were seen in sputum bacterial density or inflammatory biomarkers at 12 weeks. Aerosolized IFN-gamma1b did not improve pulmonary function, reduce sputum bacterial density, or affect inflammatory sputum markers in patients with mild-moderate lung disease.

Improving the estimation of change from baseline in a continuous outcome measure in the clinical trial setting

In many clinical trials, the primary focus is whether treatment groups differ with respect to the change from baseline to end of therapy in a continuous response variable. Randomized clinical trials often use a repeated measures design in which subjects are followed-up at fixed times throughout the study. With this design, testing for differences between treatment groups with respect to the average change from baseline to end of therapy in the response variable is equivalent to testing for differences between the rates of change in the response variable, assuming the rates of change in each treatment group are linear. This analysis can be performed quite easily using methods such as generalized estimating equations (GEE). However, if the rate of change in the response cannot be assumed linear, the average change from baseline is many times calculated using simply differences between baseline and final measurements and additional data points are not included in the analysis. Instead, we propose using all available data in a repeated measures model that is based on the nonlinear treatment response pattern to estimate the average change from baseline to end of therapy in each treatment group. GEE with robust variance estimation is used for obtaining these model-based estimates of the treatment effect and a simple test for appropriateness of the model is presented. The GEE model presented, in conjunction with the test for appropriateness of the model, form the basis for an adaptive analysis approach for determining the method of estimation of the primary endpoint. This approach results in more efficient estimates of the treatment effect when the response pattern is specified correctly and minimizes the bias in the estimate when the hypothesized response pattern is misspecified. We are motivated by examples in the cystic fibrosis (CF) clinical trial setting and demonstrate the potential for this approach in reducing the sample size required for future CF clinical trials.

Use of phage display to identify potential Pseudomonas aeruginosa gene products relevant to early cystic fibrosis airway infections

Pseudomonas aeruginosa airway infections are a major cause of morbidity and mortality in patients with cystic fibrosis. Treatment of established infections is difficult, even with microbiologically active agents. Thus, prevention of infection is an important goal of management. Isolates from cystic fibrosis patients appear to originate from the environment but adapt to the milieu of the airway of the cystic fibrosis patient and evolve toward a common phenotype. Identification of the antigens expressed early in infection may lead to novel targets for vaccine development. Immunogenic peptides were identified in a J404 random nonapeptide phage display library with serum from cystic fibrosis patients obtained within the first year of P. aeruginosa infection. One hundred sixty-five reactive clones were verified by plaque lift assays, and their inserts were sequenced. The sequenced nonapeptides were compared with the published sequence of strain PAO1, identifying homologies to 76 genes encoding outer membrane and secreted proteins. The majority of these were proteins involved in small-molecule transport, membrane structural proteins, and secreted factors. An in silico analysis was performed that suggested that the occurrence of multiple matches to predominantly outer membrane and secreted proteins was not attributable to random chance. Finally, gene expression array data from early isolates of P. aeruginosa from cystic fibrosis patients was compared with the results from phage display analysis. Eleven outer membrane and secreted proteins were common between the two data sets. These included genes involved in iron acquisition, antibiotic efflux, fimbrial biogenesis, and pyocin synthesis. These results demonstrate the feasibility and validity of this novel approach and suggest potential targets for future development.

Association between Stenotrophomonas maltophilia and lung function in cystic fibrosis

BACKGROUND

Stenotrophomonas maltophilia (SM) is a Gram-negative non-fermenting bacteria cultured from the sputum of patients with cystic fibrosis (CF). To date, no information is available regarding the effect of this organism on lung function in CF.

METHODS

A cohort study was conducted to assess the effect of SM on lung function among CF patients aged > or =6 years in the CF Foundation National Patient Registry from 1994 to 1999. Repeated measures regression was used to assess the association between SM and lung function.

RESULTS

The cohort consisted of 20 755 patients with median age at entry of 13.8 years and median follow up time of 3.8 years; 2739 patients (13%) were positive at least once for SM and 18 016 (87%) were never positive. After adjusting for sex, height and age, patients with SM had a mean forced expiratory volume in 1 second which was 0.09 l less (95% CI 0.05 to 0.14) than those without SM. The mean rate of decline associated with SM positivity was 0.025 l/year (95% CI 0.012 to 0.037) but, after adjusting for confounders (sex, height, weight, intravenous antibiotic courses, hospital admissions, pancreatic insufficiency, and Pseudomonas aeruginosa and Burkholderia cepacia status), the mean rate of decline decreased to 0.008 l/year (-0.008, 95% CI -0.019 to 0.003).

CONCLUSIONS

Although CF patients with SM have worse lung function at the time of positivity, no association was found between SM and increased rate of decline after controlling for confounders.

Inflammatory and microbiologic markers in induced sputum after intravenous antibiotics in cystic fibrosis

Induced sputum has been used to study airway inflammation. We sought to determine whether markers of infection and inflammation in induced sputum were a useful and safe outcome measure in cystic fibrosis. We hypothesized that bacterial density and inflammatory content of induced sputum would decrease after antibiotic therapy. Induced sputum was assayed for bacterial density, cell count, and differential and inflammatory markers before and after treatment with intravenous antibiotics. Fifty-five of the 72 subjects enrolled (mean age +/- SD 18.2 +/- 7.9 years) completed the study. FEV1 increased by an average 0.3 +/- 0.3 L (10.4 +/- 8.7% predicted FEV1), p<0.0001; density of Pseudomonas aeruginosa and Staphylococcus aureus decreased by 2.4 +/- 3.1 log10 cfu/g (p<0.0005) and 4.0 +/- 2.3 log10 cfu/ml (p<0.0001), respectively; neutrophil count decreased by 0.4 +/- 0.6 log10 cells/ml (p<0.0001), interleukin-8 concentration by 0.5 +/- 1.3 log10 pg/ml (p<0.05), and neutrophil elastase by 0.4 +/- 0.7 log10 microg/ml (p<0.005). Seven of 127 (6%) sputum induction procedures showed a decrease in FEV1 of 20% or more. We conclude that markers in induced sputum may be useful, noninvasive outcome measures to assess response to therapies in cystic fibrosis studies.

Azithromycin in patients with cystic fibrosis chronically infected with Pseudomonas aeruginosa: a randomized controlled trial

CONTEXT

Treatment strategies for cystic fibrosis (CF) lung disease include antibiotics, mucolytics, and anti-inflammatory therapies. Increasing evidence suggests that macrolide antibiotics might be beneficial in patients with CF.

OBJECTIVE

To determine if an association between azithromycin use and pulmonary function exists in patients with CF.

DESIGN AND SETTING

A multicenter, randomized, double-blind, placebo-controlled trial conducted from December 15, 2000, to May 2, 2002, at 23 CF care centers in the United States.

PARTICIPANTS

Of the 251 screened participants with a diagnosis of CF, 185 (74%) were randomized. Eligibility criteria included age 6 years or older, infection with Pseudomonas aeruginosa for 1 or more years, and a forced expiratory volume in 1 second (FEV1) of 30% or more. Participants were stratified by FEV1 (> or =60% predicted vs <60% predicted), weight of less than 40 kg vs 40 kg or more, and CF center.

INTERVENTION

The active group (n = 87) received 250 mg (weight <40 kg) or 500 mg (weight > or =40 kg) of oral azithromycin 3 days a week for 168 days; placebo group (n = 98) received identically packaged tablets.

MAIN OUTCOME MEASURES

Change in FEV1 from day 0 to completion of therapy at day 168 and determination of safety. Secondary outcomes included pulmonary exacerbations and weight gain.

RESULTS

The azithromycin group had a mean 0.097-L (SD, 0.26) increase in FEV1 at day 168 compared with 0.003 L (SD, 0.23) in the placebo group (mean difference, 0.094 L; 95% confidence interval [CI], 0.023-0.165; P =.009). Nausea occurred in 17% more participants in the azithromycin group (P =.01), diarrhea in 15% more (P =.009), and wheezing in 13% more (P =.007). Participants in the azithromycin group had less risk of experiencing an exacerbation than participants in the placebo group (hazard ratio, 0.65; 95% CI, 0.44-0.95; P =.03) and weighed at the end of the study an average 0.7 kg more than participants receiving placebo (95% CI, 0.1-1.4 kg; P =.02).

CONCLUSION

Azithromycin treatment was associated with improvement in clinically relevant end points and should be considered for patients with CF who are 6 years or older and chronically infected with P aeruginosa.

Susceptibility testing of Pseudomonas aeruginosa isolates and clinical response to parenteral antibiotic administration: lack of association in cystic fibrosis

STUDY OBJECTIVE

To determine the relationship between the antibiotic susceptibility of Pseudomonas aeruginosa isolated from the sputum of patients with cystic fibrosis (CF) and the patient's response to parenteral antibiotic administration, we performed a retrospective analysis using data from patients in the placebo arm of a phase 3 trial of tobramycin solution for inhalation. All patients were chronically infected with P aeruginosa. Seventy-seven of the 262 patients receiving placebo experienced a pulmonary exacerbation during the trial for which they received therapy with IV tobramycin and ceftazidime. The susceptibility of the P aeruginosa isolates to ceftazidime and tobramycin was determined at trial enrollment by broth microdilution.

DESIGN

The clinical response to combination antibiotic therapy was assessed by analyzing differences in spirometry before and after antibiotic administration. The FEV(1) percent predicted at the first visit after the conclusion of antibiotic administration was compared to the FEV(1) percent predicted prior to antibiotic therapy. The results were analyzed both descriptively and by regression analyses.

RESULTS

The conditions of 54 patients improved, and those of 9 patients worsened, and in 14 patients there was no change in FEV(1) with antibiotic administration. No correlation was observed between the susceptibility of P aeruginosa to tobramycin or ceftazidime and clinical response. Only the three following variables were observed to significantly correlate with FEV(1) after antibiotic treatment on regression analysis: FEV(1) prior to treatment (p < 0.0001); number of days elapsed between the previous FEV(1) measurement and the initiation of IV antibiotic therapy (p < 0.002); and the number of days elapsed between the determination of the minimum inhibitory concentration and the initiation of IV therapy (p < 0.03). No significant trends were observed between the antibiotic susceptibility of P aeruginosa isolates and treatment outcomes.

CONCLUSION

While lack of statistical significance for a trend between bacterial susceptibilities and the response to parenteral antibiotic administration does not mean that no such trend exists, the precision of the confidence intervals allows us to conclude that even if isolate antibiotic susceptibilities affect outcome, the impact would be small and not clinically relevant.

Developing cystic fibrosis lung transplant referral criteria using predictors of 2-year mortality

The first objective of our study was to develop a model identifying the best clinical predictors of 2-year mortality among patients with cystic fibrosis (CF), to assist in selection of appropriate candidates for lung transplantation. Using multivariate logistic regression, we found that age, height, FEV1, respiratory microbiology, number of hospitalizations for pulmonary exacerbations, and number of home intravenous antibiotic courses were all significant predictors of 2-year mortality among 14,572 patients in the Cystic Fibrosis Foundation National Patient Registry who were 6 years of age or older in 1996. The second objective was to compare the diagnostic accuracy of our model when used to guide referral for lung transplant with that of the widely used criterion of an FEV1 of less than 30% predicted. Surprisingly, this well-fitting model derived from the largest collection of data available on patients with CF provided no better diagnostic accuracy than the simpler FEV1 criterion. Both had high negative predictive values (98 and 97%, respectively) but only modest positive predictive values (33 and 28%, respectively). Transplant referral decisions based either on a multivariate logistic model or on the criterion of an FEV1 of less than 30% predicted are likely to result in high rates of premature referral. Better clinical predictors of short-term mortality among patients with CF are needed.

The cystic fibrosis therapeutics development network (CF TDN): a paradigm of a clinical trials network for genetic and orphan diseases

Clinical trials have become critical to the advancement of medical science and to the evolution of patient care in medicine. The science of clinical research has advanced from early studies in which treatment was assessed without controls to sophisticated multinational collaborative randomized, double-blind, placebo controlled trials of therapeutic interventions. To facilitate the advancement of clinical research, clinical trials networks have been developed to conduct multicenter studies. This review describes the history of clinical trials, clinical trials networks, and the goals of such networks in the United States. The Cystic Fibrosis Therapeutics Development Network, a network that represents the paradigm for genetic and orphan diseases, is described in detail. This network has been extremely successful in its first 3.5 years of existence conducting 18 different clinical trials in patients with Cystic Fibrosis. Unique aspects of the network include the use of internet applications for study conduct and communication, the development of statistical methodology to enhance the efficiency of clinical trial design, the development of outcome measures specific to Cystic Fibrosis, and the development of infrastructure necessary for expediting protocol development. In the current environment, clinical research faces significant challenges related to ensuring the safe and ethical conduct of clinical research while promoting fast and efficient clinical trials. To succeed and move forward to provide treatments and find cures for diseases, clinical trials networks must continue to evolve. The Cystic Fibrosis Therapeutics Development Network represents a network that has met this challenge and will continue to provide a venue for the safe and efficient conduct of clinical trials in Cystic Fibrosis.

Correlation between an in vitro invasion assay and a murine model of Burkholderia cepacia lung infection

Our understanding of the virulence of Burkholderia cepacia complex lung infections in cystic fibrosis patients is incomplete. There is a great deal of variability in the clinical course, from simple colonization to severe and often fatal necrotizing pneumonia, termed cepacia syndrome. Multiple subspecies (called genomovars) have been identified, and these genomovars may hold the key to understanding the variable pathogenicity. Thirty-one B. cepacia complex isolates belonging to five of the seven genomovars were examined by using a gentamicin protection assay of invasion with A549 cells. The level of epithelial cell invasion by B. cepacia in the A549 model was relatively low compared with the data obtained for other pathogens and was often variable from assay to assay. Thus, a statistical approach was used to determine invasiveness. When this model was used, one of four genomovar I strains (25%), three of eight genomovar II strains (37.5%), seven of nine genomovar III strains (77.8%), one of four genomovar IV strains (25%), and none of the four genomovar V strains examined were defined as invasive. All other strains were categorized as either noninvasive or indeterminate. Invasive, noninvasive, and indeterminate isolates belonging to genomovars II and III were subsequently tested for splenic invasion with the mouse agar bead model. Correlation between the models for six strains was demonstrated. Our results indicate that a statistical model used to determine invasiveness in an in vitro invasion assay can be used to predict in vivo invasiveness.

A regression modeling approach for describing patterns of HIV genetic variation

We introduce a novel approach for describing patterns of HIV genetic variation using regression modeling techniques. Parameters are defined for describing genetic variation within and between viral populations by generalizing Simpson's index of diversity. Regression models are specified for these variation parameters and the generalized estimating equation framework is used for estimating both the regression parameters and their corresponding variances. Conditions are described under which the usual asymptotic approximations to the distribution of the estimators are met. This approach provides a formal statistical framework for testing hypotheses regarding the changing patterns of HIV genetic variation over time within an infected patient. The application of these methods for testing biologically relevant hypotheses concerning HIV genetic variation is demonstrated in an example using sequence data from a subset of patients from the Multicenter AIDS Cohort Study.