A Multi-Dimensional Signature to Predict Treatment Failure in Patients with Connective Tissue Disease-Associated Interstitial Lung Disease - PROJECT SUMMARY Interstitial lung disease (ILD) is a frequent and often fatal complication of connective tissue disease (CTD). While immunosuppression constitutes first-line therapy for most CTD-ILD subtypes, many patients experience treatment failure, experiencing relentless lung function decline and early death despite therapy. Unfortunately, the clinical trajectory of CTD-ILD patients receiving immunosuppressant therapy is incompletely understood and there are no tools to predict treatment failure. Reliably predicting treatment failure would allow for consideration of stronger upfront immunosuppressant therapy, early initiation of antifibrotic therapy, and timely referral for lung transplantation. Tools to predict treatment failure are urgently needed. Dr. Pugashetti is a Pulmonary and Critical Care physician at the University of Michigan and will be principal investigator for this mentored K23 project. Her long-term goal is to become a leader in developing predictive biomarkers to guide therapeutic selection, thereby improving outcomes for patients with CTD-ILD. The objective in this application is to develop predictive biomarkers of treatment failure, defined as 12-month forced vital capacity decline, death, or lung transplant while treated with immunosuppressant therapy. In Aim 1, FVC trajectory and cumulative incidence of treatment failure across CTD-ILD subtypes will be determined. For Aim 2, retrospective multi-center proteomic data were used to generate an exciting preliminary signature that effectively discriminated treatment failure from immunosuppressant response/stability. In Aim 2, Dr. Pugashetti will optimize and prospectively validate this proteomic signature and compare test performance to a clinical prediction model. Lastly, in Aim 3, radiomic variables will be tested for association with treatment failure and will be added to a clinical and proteomic model to determine whether they augment treatment prediction. Findings from all three aims will be validated prospectively. Dr. Pugashetti’s skills gap-informed training plan consists of formal mentorship, didactic coursework, scientific conferences, and progressively independent research activities. She is ideally positioned to successfully complete this work given her prior research training, career development plan, outstanding institutional support from the University of Michigan, and highly committed mentoring team with expertise in ILD and proteomics (Oldham), CTD-ILD (Khanna), advanced biostatistical methods (Murray), radiomics (Kazerooni) and pulmonary patient-oriented research (Han). By the end of the award period, she will develop expertise at the intersection of ILD, rheumatology, and biomarker investigation and develop the skills to become an independent physician- scientist. Completion of this project will lead to an R01 to assemble a multi-center, prospective CTD-ILD cohort and investigate longitudinal changes in biomarkers of treatment failure.
