Early posttransplant reductions in club cell secretory protein associate with future risk for chronic allograft dysfunction in lung recipients: results from a multicenter study

Abstract

Background: Chronic lung allograft dysfunction (CLAD) increases morbidity and mortality for lung transplant recipients. Club cell secretory protein (CCSP), produced by airway club cells, is reduced in the bronchoalveolar lavage fluid (BALF) of lung recipients with CLAD. We sought to understand the relationship between BALF CCSP and early posttransplant allograft injury and determine if early posttransplant BALF CCSP reductions indicate later CLAD risk.

Methods: We quantified CCSP and total protein in 1606 BALF samples collected over the first posttransplant year from 392 adult lung recipients at 5 centers. Generalized estimating equation models were used to examine the correlation of allograft histology or infection events with protein-normalized BALF CCSP. We performed multivariable Cox regression to determine the association between a time-dependent binary indicator of normalized BALF CCSP level below the median in the first posttransplant year and development of probable CLAD.

Results: Normalized BALF CCSP concentrations were 19% to 48% lower among samples corresponding to histological allograft injury as compared with healthy samples. Patients who experienced any occurrence of a normalized BALF CCSP level below the median over the first posttransplant year had a significant increase in probable CLAD risk independent of other factors previously linked to CLAD (adjusted hazard ratio 1.95; p = 0.035).

Conclusions: We discovered a threshold for reduced BALF CCSP to discriminate future CLAD risk; supporting the utility of BALF CCSP as a tool for early posttransplant risk stratification. Additionally, our finding that low CCSP associates with future CLAD underscores a role for club cell injury in CLAD pathobiology.

Keywords: CLAD risk; bronchoalveolar lavage fluid; chronic lung allograft dysfunction; club cell secretory protein; lung transplantation.

Figure 1.

Patient and sample cohort for analysis *Healthy samples were defined as those where the biopsy did not show any rejection or non-rejection histology and there was no infection; AR/LB only samples were defined as those where only histological AR and/or LB was present in the absence of any non-rejection histology or infection; AR/LB + other histology or infection samples were defined as those where AR and/or LB was present along with other findings such as non-rejection histology or infection; OP/ALI only samples were defined as those where only histological OP and/or ALI was present in the absence of any rejection histology or infection; OP/ALI + other histology or infection samples were defined as those where OP and/or ALI was present along with other findings such as rejection histology or infection; Infection only was defined as the presence of any bacterial, fungal, mycobacterial, or viral organisms in the BALF in the absence of any rejection or non-rejection histology on biopsy; Infection + Other groups are defined as detection of the specified type of organism along with other findings on biopsy such as rejection or non-rejection histology. Some samples may be represented in more than one group.

Figure 2.

Association between BALF sample groupings of interest and protein-normalized BALF CCSP concentration, adjusted for time posttransplant, donor age, and transplant type (single versus bilateral).

Figure 3.

An extended Kaplan-Meier estimator that allows for time-varying covariates was used to describe the cumulative incidence of probable CLAD stratified by whether the patient had (red line) vs. did not have (blue line) at least one occurrence of a reduced protein-normalized BALF CCSP concentration (defined as below the median) in the first posttransplant year.