PRS3 Joint Modelling of Forced Vital Capacity Decline and Time to Hospitalization Outcomes in Systemic Sclerosis-Associated Interstitial Lung Disease

Abstract

Joint modelling can be used to determine whether a biomarker can serve as a surrogate for a clinically important endpoint that may otherwise be difficult to measure, which is of particular relevance in rare diseases. We used this approach to assess the validity of forced vital capacity (FVC) decline as a surrogate endpoint for hospitalization-associated endpoints in systemic sclerosis-associated interstitial lung disease (SSc-ILD). Joint models for longitudinal and time-to-event data were applied to data from the Phase III SENSCIS® trial, which assessed the efficacy and safety of nintedanib in patients with SSc-ILD during the treatment period over 52 weeks. A joint model was fitted for each endpoint (time to first: "all-cause hospitalization or death”; “SSc-related hospitalization or death”; and “admission to emergency room [ER] or hospital followed by admission to intensive care unit [ICU] or death”). The “estimated slope” of FVC% predicted was the shared parameter, using all FVC values recorded during the SENSCIS® trial prior to hospitalization events. Independent implementations of the methodology with an available SAS macro and an R package were applied to validate the results. FVC decline had a statistically significant impact on the risk of all-cause and SSc-related hospitalizations or death (both P<0.0001). An FVC decrease of 5% corresponded with a 1.81-fold (95% confidence interval [CI] of hazard ratio [HR]: 1.42, 2.30) and 1.91-fold (95% CI of HR: 1.41, 2.60) increase in risk of “all-cause hospitalization or death” and “SSc-related hospitalization or death”, respectively. However, FVC decline did not impact the risk of “admission to ER or hospital followed by admission to ICU or death” (P=0.15). We found a consistent association between FVC decline and hospitalization-associated endpoints in patients with SSc-ILD. Our findings provide evidence that FVC decline may serve as a surrogate endpoint for clinically important hospitalization-associated endpoints.