Dynamics of human regulatory T cells in lung lavages of lung transplant recipients

Abstract

Background: Despite advances in the field of lung transplantation, the median survival after lung transplant remains below 5 years. Early rejection is a risk factor for the development of chronic rejection. In animal models of transplant tolerance, regulatory T cells (Tregs) can prevent the establishment of rejection.

Methods: This study was designed to explore the dynamics of Tregs focally and systemically in lung transplant recipients. Sequential surveillance bronchoscopy results were available in 51 patients with at least four sequential samples recovered from each patient at defined times posttransplant. In 36 individuals, a complete year of follow-up data for BAL was analyzed. In 33 of these individuals had a complete year of follow-up data for peripheral blood monocyte cell specimens were also analyzed. Lung lavage cells were recovered from each bronchoscopy and corresponding blood draw and subjected to polychromatic flow cytometry. The percentage of CD4 lymphocytes, which expressed the intracellular transcription factor FoxP3 was recorded at each point. At each time point, lung biopsy specimens were scored for rejection.

Results: Lung Treg frequency was significantly more variable than blood Treg frequency. Treg frequency in the lung was increased in the aftermath of acute rejection. In contrast, lung Treg frequency declined sequentially in patients demonstrating continued quiescence. Mean BAL Treg level integrated over the first transplant year correlated inversely with the degree of acute cellular rejection. In contrast, blood Treg levels demonstrated no correlation with lung pathology.

Conclusions: Lung Tregs increase in the setting of acute cellular rejection, whereas declining levels of BAL Tregs correlates with immunologic quiescence.

Fig 1. BAL Tregs increase after early rejection and decrease in the setting of back-to-back quiescence

12 patients demonstrated acute cellular rejection on the first biopsy (left panel). In subsequent biopsies there was a trend toward increased BAL tregs (p=.0586 between 2 weeks and 1 month). 20 patients demonstrated no rejection on the first 3 surveillance biopsies. There was a significant drop in BAL Tregs over that period p<.001 between 2 weeks and 1 month. The mean frequency of each time point is indicated in the boxes.

Fig 2. BAL Tregs increase in the aftermath of untreated rejection

There were 22 instances of rejection which was not treated among the 52 individuals studied. In 16 of these episodes there was a subsequent increase in BAL Tregs recovered on the next bronchoscopy. Overall there was a 3.4% increase in BAL Treg frequency in the aftermath of untreated rejection (p<.005 by paired t-test).

Fig 3. Tregs are consistently present in BAL but do not vary significantly with the time post transplant

Grey boxes show the frequency of BAL Tregs from the first 4 sequential biospies performed at 2 weeks, 1 month, 2 month, and 3month post transplant. Numbers below the box represent the mean Treg frequency. The solid line within the box represents the median frequency. The number above each box indicates mean frequency. P=ns between all time points by 2-tailed t-test. Hatched boxes show Treg frequency in the clinical conditions of the first episode of rejection, the follow-up after rejection and before rejection. P=.0016 between before and after rejection by 2-tailed t-test.

Fig 4. PBMC Treg frequency does not vary over time or by clinical condition

Grey boxes show the frequency of pbmc Tregs isolated at the time of the first 4 sequential biospies post transplant. Numbers below the box represent the mean Treg frequency. The solid line within the box represents the median frequency. The number above each box indicates mean frequency. P=ns between all time points by 2-tailed t-test. Hatched boxes show Treg frequency in the clinical conditions of the first episode of rejection, the follow-up after rejection and before rejection. P= 0.20 between before and after rejection by 2-tailed t-test.

Fig 5. BAL but not pbmc Treg frequency correlates with rejection score in a longitudinal assessment

For all patients with a year of follow-up and at least 5 BAL specimens and at least 4 pbmc specimens the mean Treg frequency was calculated. The cumulative rejection score was calculated as the total A-score and B-score for all transbronchial biopsies performed in the first year. There was a strong correlation between mean BAL treg frequency and rejection score (R²= .44, p<.001). There was no correlation between mean pbmc Treg frequency and rejection score (R²=.0005, p=0.8).

Fig 6. Longitudinal patterns of BAL and pbmc Tregs in individual patients

Solid line Indicate BAL Treg frequency and dassed line indicates pbmc Treg frequency. Top panel Shows pattern in individual with no rejection in the first post transplant year. Middle panel Shows pattern in a patient with 2 episodes of early minimal rejection, followed by a Period of prolong quiescence. Bottom panel shows pattern in patient with back-to-back Rejection starting at 6 months post transplant.