Inter- and intrasubject variability of the inflammatory response to segmental endotoxin challenge in healthy volunteers.

Segmental endotoxin challenge with lipopolysaccharide (LPS) can be used as a pharmacodynamic model to safely induce a transient airway inflammation in the peripheral lung of healthy subjects and to test the anti-inflammatory efficacy of investigational new drugs. In contrast to whole lung LPS challenge only a fraction of the dose is required that can be precisely administered to a specific lung region and a vehicle challenged segment as an intra-subject control can be included. The aim of this study was to assess the intra- and inter-individual variability of the response to segmental LPS challenge for the appropriate design and power calculation of future clinical trials. Two cohorts with 10 subjects each underwent two segmental LPS challenges within five weeks. The inflammatory response was evaluated in bronchoalveolar lavage (BAL) fluid at 6 (cohort 1) and 24 h (cohort 2) both in the LPS and in a vehicle challenged segment, as well as in plasma for up to 26 h post LPS challenge. While the cytokine response was more pronounced at 6 h, the influx of neutrophils and monocytes dominated at 24 h; e.g. neutrophils increased from a median (inter-quartile range, IQR) of 0.14 (0.16) and 0.09 (0.08)x10(4) cells/mL BAL fluid at baseline to 10.2 (17.1) and 19.3 (15.9)x10(4) cells/mL 24 h after the two separate challenges. The within-subject variability was higher than the between-subject variability for most of the markers. However, sample size estimations based on the variability of outcome variables found lower or equal numbers with cross-over designs compared to parallel group designs for cellular markers at 24 h and cytokine variables at 6 h. The segmental LPS challenge model was safe. Future study designs have to balance between burden to the study subjects (4 versus 2 bronchoscopies), variability (within-versus between-subject), and the desired outcome variable (cells versus chemo/cytokine).