Validation of the Lung Function Score as a Valuable Parameter for the Assessment of Chronic Graft-Versus-Host Disease and Clinical Outcome After Allogeneic Haematopoietic Cell Transplantation.

Abstract 2259

Poster Board II-236

The modified lung function score (LFS), which combines changes in forced expiratory volume in the first second (FEV1) and the capacity of the lung for carbon monoxide adjusted for haemoglobin level (cDLCO), has been proposed by the NIH consensus project on criteria for clinical trials in chronic graft versus host disease (cGVHD) in 2006, but since then it has not been validated. We evaluated the outcome of 205 patients who underwent allogeneic haematopoietic cell transplantation (allo-HCT) between 1998 and 2006 with regard to their LFS (grade I to IV) before HCT as well as to their LFS at 90 days and at 1 year after HCT, and compared the impact of LFS with standard grading systems for changes in FEV1, cDLCO, total lung capacity (TLC) and forced vital capacity (FVC). Mean follow-up was 973.3 (22 – 3091) days.

Before allo-HCT, 35.2% of patients had a LFS I, 58.4% a LFS II, 6.4% a LFS III and none a LFS IV, respectively. We found no correlation of overall survival (OS) with pre-transplantation-FEV1, -TLC or -FVC, whereas a significant association was seen with the cDLCO (5-year OS: cDLCO > 80%: 50.7% vs. cDLCO 70-80%: 34.1% vs. cDLCO 60-70%: 56.3% vs. cDLCO 50-60%: 41.3% vs. cDLCO < 50%: 20.0%, p = 0.045). A higher pre-transplantation-LFS did show a trend towards increased mortality but did not reach statistical significance (5-year OS LFS I: 51.4% vs. LSF II: 44.1% vs. LFS III: 30.8% vs. LFS IV: NA; LFS I vs. LFS III: p = 0.057).

A reduction in lung function parameters after allo-HCT is often associated with the development of pulmonary graft versus host disease. According to the current ATS/ERS guidelines, FEV1 can be used as universal parameter for grading the severity of obstructive as well as restrictive pulmonary function changes. Prior studies demonstrated, that FEV1 decreases frequently after allo-HCT, and that cDLCO is often reduced before allo-HCT and shortly after with subsequent increase or recovery. Respective changes where also seen in our patient cohort. A reduction in FEV1 at 3 months and at 1 year after allo-HCT did strongly correlate with increased overall mortality (5-year OS (FEV1 on day 90): FEV1 > 80%: 72.1% vs. FEV1 70-80%: 50.0% vs. FEV1 60-70%: 0% vs. FEV1 50-60%: 25.0% vs. FEV1 <50%: NA, p < 0.001; 5-year OS (FEV1 on day 360): FEV1 > 80%: 82.7% vs. FEV1 70-80%: 69.3% vs. FEV1 60-70%: 83.3% vs. FEV1 50-60%: NA vs. FEV1 < 50%: 33.3% p = 0.012), whereas changes of the cDLCO only at day 360 (5-year OS: cDLCO > 80%: 82.8% vs. cDLCO 70-80%: 80.9% vs. cDLCO 60-70%: 80.1% vs. cDLCO 50-60%: 84.6% vs. cDLCO < 50%: 60.0%, p = 0.02) but not at day 90 (cDLCO > 80%: 53.5% vs. cDLCO 70-80%: 80.6% vs. cDLCO 60-70%: 73.1% vs. cDLCO 50-60%: 56.1% vs. cDLCO < 50%: 60.0%, p = 0.11) could be associated with differences in survival. A LFS of I/II/III/IV was found in 22.5/64.0/12.6/0.9% at day 90 and in 24.5/63.3/11.2/1.0% at day 360, respectively, and a clear correlation was seen between the development of cGvHD and elevated LFS levels at day 360 (no cGvHD: 20.0% LFS I, 76.7% LFS II, none LFS III, 3.3% LFS IV vs. limited cGvHD: 27.6% LFS I, 65.5% LFS II, 6.9% LFS III, none LFS IV vs. extensive cGvHD: 25.8% LFS I, 48.4% LFS II, 25.8% LFS III, none LFS IV, p = 0.004). As for FEV1, both at day 90 and 360 after allo-HCT there was a clear correlation between the level of LFS and overall mortality (5-year OS: LFS I: 69.5%, LFS II: 70.4%, LFS III: 45.1%, LFS IV: 0% at day 90 and LFS I: 90.2%, LFS II: 81.6%, LFS III: 64.9%, LFS IV: 0% at day 360, p < 0.001).

In summary, the LFS presents a valid approach to assess and clinically interpret changes in lung function in patients undergoing allo-HCT. Our data show, that pre-HCT LFS scores potentially provide information on allo-HCT outcome, allowing a better risk assessment of transplantation-related mortality. More importantly, post-HCT LFS levels strongly correlate with cGvHD and overall survival, making the LFS a valuable tool for adequate and tailored patient follow-up and adaption of immunosuppression. LFS assessment should therefore be routinely incorporated in the pre- and postransplantation care of allo-HCT patients.