Background/Purpose:

Scleroderma related interstitial lung disease (SLD) is a major cause of morbidity and mortality in severe systemic sclerosis (SSc). The Scleroderma: Cyclophosphamide or Transplantation (SCOT) trial) demonstrated that myeloablation followed by autologous hematopoietic stem cell transplant (HSCT) significantly improved event-free and overall survival of SSc patients at 54 months compared with 12 monthly treatments with intravenous cyclophosphamide (CYC) (Sullivan KM, et al, N Engl J Med 2018;378:35-47). The aim of the present study was to follow the changes in lung parenchymal abnormalities between baseline and serial follow-up high-resolution CT (HRCT) scans performed in SCOT participants at least yearly for up to 5 years.

Methods: Quantitative scores of SLD were measured using computer based quantitative image analysis of standardized non-contrast volumetric thin section thoracic HRCT. The same CT machine was used for all time points (except for one subject) with careful attention to breath hold reproducibility and image quality. Quantitative CT texture-based scores of disease-extent including Quantitative interstitial lung disease (QILD) and quantitative lung fibrosis (QLF) were derived from a previously described supervised texture classification model (Kim et al Clin Exp Rheumatol, 28 (5 Suppl 62) (2010), S26–S35). Mixed effect models with an interaction between treatment arms and duration were used to compare the changes from the baseline in QLF and QILD scores in whole lung (WL) and the most severe lobe (MSL). Spearman rank correlations were used to test the association between the changes in QLF and QILD versus pulmonary function tests (PFT) in the overall study population

Results: All 75 randomized subjects had baseline lung CT studies. Quantitative scores were calculated for subjects with available follow-up HRCT scans at 14, 26, 48, and 54 months. Baseline characteristics were not different between the two groups. WL QILD scores decreased significantly for the HSCT vs no change in CYC groups (p=0.024). There was a significant difference in WL increasing QLF in CYC vs stability in the HSCT arm (p=0.047). Changes in means (±SE) at 54 months in treatment completers were -7%(±2) for HSCT and 0%(±5) for CYC in QILD, and -1%(±1) for HSCT and +3%(±3) for CYC in QLF. Similarly, significant differences in treatments were also found in MSL (p=0.004 in QILD and p=0.002 in QLF). The direction of change in structural measures of QILD and QLF for both WL and MSL tracked with physiologic PFT measures in the overall study population as shown in the table

Conclusion:

Changes in quantitative lung CT scores of SLD provide independent validation of benefit of HSCT compared to CYC in severe SSc. Imaging improvements after HSCT continue for up to 54 months after randomization providing radiologic confirmation of long-term benefit.

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/changes-in-quantitative-scleroderma-lung-ct-measures-in-patients-treated-with-cyclophosphamide-or-transplantation/