Lung allograft dysfunction correlates with gamma-interferon gene expression in bronchoalveolar lavage

Preceding episodes of acute cellular rejection (ACR) may predispose lung allografts to the subsequent development of irreversible dysfunction or bronchiolitis obliterans syndrome (BOS). Other histologic patterns such as bronchiolitis obliterans with organizing pneumonia (BOOP), organizing pneumonia, lymphocytic bronchiolitis and diffuse alveolar damage (DAD) may also adversely affect allograft function. We have previously reported the predominant expression of Th1 cytokines (IL-2 and interferon gamma) in rejecting and Th2 (IL-10) in a tolerant model of rat lung transplantation. Here we correlate the "Th1/Th2 paradigm" in clinical lung transplantation with histologic findings and assess the effect on serial spirometric function.

We examined the mRNA expression of IL-2, interferon gamma, IL-10 and ICAM-1 in 53 bronchoalveolar lavage (BAL) specimens from 23 lung transplant (LT) recipients utilizing qualitative "nested" reverse transcriptase polymerase chain reaction (RT-PCR). We also measured IgG1 and IgG2 levels in 44 BAL specimens by ELISA. The mRNA expression for cytokines, ICAM-1 and the IgG2/IgG1 ratios were correlated with the presence or absence of ACR and alternate "histologic patterns". Serial spirometry were analyzed for the 2-3 month interval before bronchoscopic (FOB) assessment to derive "baseline" forced expiratory volume-one second (FEV1) values. The change in FEV1 coincident with (deltaFEV1 pre) and for the 2-3 month interval subsequent to (deltaFEV1 post) FOB were expressed relative to "baseline" spirometric indexes.

Detection of mRNA for interferon gamma and ICAM-1 correlated significantly with ACR, whereas IL-2 and IL-10 expression did not correlate. IL-10 was virtually "ubiquitous" in most BAL samples irrespective of the presence or absence of ACR. The highest correlation was observed with interferon gamma for acute cellular rejection whereupon the sensitivity was 77.7%, specificity 87.7%, positive predictive value 73.6% and negative predictive value 88.2%, although for ICAM-1 these values were 75%, 65.7%, 50.0% and 85.0%, respectively. Nevertheless, 4 of 5 episodes of respiratory tract infection (bacterial, CMV, Aspergillus spp.) were similarly associated with cytokine mRNA. The ratios of IgG2 to IgG1, a reflection of Th1/Th2 influence, were not statistically different when analyzed for the presence or absence of ACR (0.91+/-0.53 vs. 1.02+/-0.70, respectively; p = NS). By analysis of FEV1 trends, expression of interferon gamma was associated with a greater and persistent decrement (deltaFEV1 pre: -0.265+/-0.78 liters, and post: -0.236+/-0.1161; mean +/- SE) than ACR in the absence of interferon gamma expression (+0.158 +/- +0.065 and +0.236+/-0.007 liters, respectively) (Student-Newman-Keuls, p<.05).

Our findings suggest that interferon gamma mRNA expression and ICAM-1 may be valuable in both the diagnosis and prognosis for lung allograft ACR. IL-10, a Th2 cytokine, was locally expressed both in the presence and absence of ACR. Expression of mRNA for interferon y in BAL and, to a lesser extent ICAM-1, were associated with increased lung allograft dysfunction. Whether BAL cytokine "immunosurveillance" would complement or possibly supplant a specific "histologic pattern" and thereby direct different therapies after lung transplantation, may be potentially rewarding areas of further investigation.