Modified GAP index by including serum KL-6 to assess risk of disease progression in patients with interstitial lung diseases (ILD): preliminary results from the VAMOS study

Introduction: The development of a progressive fibrotic phenotype of ILD (PF-ILD) is still unpredictable. GAP (gender, age and physiology) index is a validated tool to predict mortality in ILD, but few is known about its association with disease progression. Serum KL-6, a lung epithelial mucin, is an established marker of pulmonary fibrosis. Objective: To investigate whether enriching GAP index with KL-6 could lead to better stratification of ILDs according to progression risk. Methods: ILD patients from 3 European centres, enrolled in the VAMOS study, were included. Disease progression was defined as relative decline ⩾10% in FVC over 12 months. Serum KL-6 was measured using automated immunoassay (Fujirebio). A stepwise logistic regression was used to select and weigh variables to be included in the new model. Results: 140 patients (32% IPF, 64% non-IPF, 4% unclassified) were included, 45 of them had PF. Stratification based on original GAP was 51%, 42% and 6% for stage I, II and III respectively. Five predictors with odds ratio>1.2 at logistic regression were included in the new GAP index: age, male, BMI, FVC, IPF and KL-6. The final model had a ROC AUC=0.695 [0.603-0.787] vs original GAP AUC =0.561 [0.460-0.663] (p =0.002) with a 58% sensitivity and 74% specificity to predict disease progression. Modified GAP index allowed to cluster patients who had PF into higher risk stage compared to the original GAP: 28/14/58% (p=0.001) vs 44/49/7% (p=0.43) in stage I, II and III respectively. Conclusions: Modified GAP index adding KL-6 and BMI can provide a better risk stratification for PF in ILD.