Predictive Value of Flow Cytometry Quantification of BAL Lymphocytes and Neutrophils in ILD

Abstract

Interstitial lung diseases (ILDs) are pathologies affecting the pulmonary interstitium and, less frequently, the alveolar and vascular epithelia. Bronchoalveolar lavage (BAL) is commonly used in ILD evaluation since it allows the sampling of the lower respiratory tract. The prognostic value of BAL cell counts in ILD is unknown. Flow cytometry quantification of lymphocytes and neutrophils in BAL of 1074 real-life consecutive patients were retrospectively correlated with clinical, radiological, anatomopathological, functional/spirometry, and evolutionary data. Cut-offs with predictive value were established at 7% and 5% for lymphocytes and neutrophils, respectively. Three risk stratification groups (Risk-LN) were established: FAVORABLE (lymphocytes > 7% and neutrophils < 5%), INTERMEDIATE (rest of patients), and UNFAVORABLE (lymphocytes < 7% and neutrophils > 5%), showing 75th percentile overall survival (OS) of 10.0 ± 1.4, 5.8 ± 0.6, and 3.0 ± 0.3 years (p < 0.001), respectively. A scoring model combining Risk-LN and the age of the patients with great predictive capacity for OS on fibrotic and non-fibrotic ILDs is proposed. This score is an independent predictive factor (HR = 1.859, p = 0.002) complementary to the fibrosis status (HR = 2.081, p < 0.001) and the type of treatment. Flow cytometry of BAL provides rapid and accurate quantification of lymphocytes and neutrophils, allowing the establishment of a risk score model that is useful in the clinical management of fibrotic and non-fibrotic ILDs from the time of diagnosis.

Keywords: bronchoalveolar lavage; flow cytometry; lung diseases; lymphocyte; neutrophil; stratification.

Figure 1

Flow cytometry analysis of the leukocyte subsets contained in BAL samples. Leukocyte subsets were identified following the hierarchical and logical gating strategy shown in the figure. TruCount beads were used to calculate the absolute cell counts per microliter following the manufacturer’s instructions.

Figure 2

Risk stratification of ILD and infectious disease based on lymphocyte and neutrophil counts of BAL. (a) Receiver operating characteristic curve (ROC) of lymphocytes and neutrophils related to overall survival (OS). Area under the curve (AUC), cut-off, and the highest sensitivity and specificity are shown. (b) Kaplan–Meier and Log-Rank tests for OS of ILD and infectious disease patients with lymphocytes and neutrophils below and over their respective cut-offs and OS of general population (without lung diseases). (c) Kaplan–Meier and Log-Rank tests for OS of ILD patients according to the combinations of lymphocytes (above or below 7%) and neutrophils (above or below 5%). (d) Kaplan–Meier and Log-Rank tests for OS of ILD patients according to the Risk-Lymphocyte/Neutrophil stratification (Risk-LN): favorable (lymphocyte > 7% and neutrophil < 5%), unfavorable (lymphocyte < 7% and neutrophil > 5%), and intermediate (rest). The 75th percentile OS (Mean ± SD) are shown for the patients and general population in each section.

Figure 3

Risk-Lymphocyte/Neutrophil (Risk-LN) stratification offers predictive information in both ILDs and infectious diseases. Kaplan–Meier and Log-Rank tests for overall survival (OS) according to Risk-LN in patients with ILD pathologies and in patients with infectious disease. The 75th-percentile-OS is shown.

Figure 4

Distribution of Risk-Lymphocyte/Neutrophil (Risk-LN) stratification groups among pulmonary pathologies. (A) Percentage of lymphocytes and neutrophils in the flow cytometry analysis of BAL samples from patients with different pulmonary pathologies. * p < 0.05, ** p < 0.01, *** p < 0.001 in the ANOVA and DMS post hoc tests. The dashed lines indicate the mean value of lymphocytes (6.5%) and neutrophils (2.12%) in the control BAL group. (B) Distribution of Risk-LN groups among pulmonary pathologies: favorable (lymphocyte > 7% and neutrophil < 5%), unfavorable (lymphocyte < 7% and neutrophil > 5%), and intermediate (rest).

Figure 5

Risk-Lymphocyte/Neutrophil stratification (Risk-LN) in ILD patients with pulmonary fibrosis. (a) Frequency of the most common lung imaging patterns among ILD subtypes. (b) Kaplan–Meier and Log-Rank tests for overall survival (OS) of ILD patients according to their predominant lung pattern. * other radiological patterns. (c) Kaplan–Meier and Log-Rank tests for OS of ILD patients with or without fibrosis according to Risk-LN stratification. The 75th-percentile-OS is shown in each case.

Figure 6

Risk-Lymphocyte/Neutrophil (Risk-LN) stratification in ILD patients with different systemic immunosuppressive treatments. (a) Kaplan–Meier and Log-Rank tests for overall survival (OS) of ILD patients according to the type of treatment: no systemic treatment, corticosteroids, or other immunosuppressants (rituximab, azathioprine, mycophenolate mofetil, cyclophosphamide, tacrolimus). (b) Kaplan–Meier and Log-Rank tests for OS of ILD patients according to the Risk-Lymphocyte/Neutrophil stratification (Risk-LN) and the type of treatment. The 75th-percentile-OS is shown in each case.

Figure 7

Overall survival risk scoring systems in ILDs. (a) Kaplan–Meier and Log-Rank tests for overall survival (OS) of ILD patients according to a scoring model including age range (<55 years = 0, 55–70 years = 1, >55 years = 2) and Risk-Lymphocyte/Neutrophil stratification (Risk-LN) (favorable = 0, intermediate = 1, unfavorable = 2). The 75th-percentile-OS was estimated separately for patients with or without pulmonary fibrosis. (b) Cox regression analysis for OS in ILD patients for sex, type of ILD, smoking status, type of treatment, fibrosis status, and the OS risk score.