B101-03 Quantitative Computed Tomography Identifies BOS in Allogeneic Hematopoietic Cell Transplant Recipients. A Multicenter Prospective Trial

Rationale Advances in imaging analysis, including quantitative computed tomography (qCT) have shown utility in patients with bronchiolitis obliterans syndrome (BOS) following hematopoietic cell transplantation (HCT). A prospective, multicenter trial of qCT metrics in patients with newly diagnosed BOS was conducted with the aim of differentiating BOS from those with chronic graft-versus host disease (cGVHD) without lung disease, and to characterize disease physiology. Methods Patients with newly diagnosed or acutely worsening cGVHD (n = 24, cohort 1), or BOS (n = 38, cohort 2) were enrolled prospectively in an IRB approved trial at six academic centers, with enrollment currently on-going. NIH consensus criteria were used to define BOS. Parametric response mapping (PRM), airway and vessel wall measurements were quantitated using biphasic CT images. PRM metrics of functional small airway disease (fSAD), parenchymal lung disease (PD), emphysematous changes (EM) and normal lung attenuation (Norm) were measured as a % of total lung volume and correlated with concurrent pulmonary function testing (PFT). Airway and vessel wall measurements, including “wall area percent” (WA%), Pi10, bronchiectasis indices (BEI), vessel volume and indirect measures of ventilation and regional/global lung mechanics (Jacobian determinants) and blood perfusion (dBlood) were examined. Results The mean forced exhaled volume in one second (FEV1), forced vital capacity (FVC) and forced expiratory flow at 25-75% (FEF25-75) were each significantly lower in patients with BOS, when compared to patients with chronic GVHD, but without BOS (p < 0.001), TABLE 1. Likewise, the mean PRM fSAD was significantly higher in patients with newly diagnosed BOS versus those without BOS, 24±17% vs 8 ±11% (p < 0.01). In contrast, no differences in vessel and airway wall measurements were seen between the two populations, including the WA%, BEI, Pi10 (mean airway thickness normalized to a theoretical airway of 10 mm inner perimeter), and a subsegmental wall thickness score (BV5). The Jacobian mean and dBlood were significantly lower (indicating loss of ventilation and perfusion) in patients with BOS than those chronic GVHD alone (<0.01). Conclusion CT metrics could differentiate patients with newly diagnosed or acutely worsening BOS post-HCT from those with other cGVHD. Our data validate that PRM fSAD is significantly increased when compared to patients with chronic GVHD without BOS. However, no differences in large airway or vessel wall thickness were seen between the cohorts. Our data further suggest that early or progressive airway remodeling in patients with BOS is occurring at the level of the terminal bronchiole, and not in larger airways.