Secondary Analyses to Support the Rational Design of Clinical Trials in Chronic Lung Allograft Dysfunction

点击翻译按钮获取中文摘要

Long-term outcomes for lung transplant recipients remain inadequate. Chronic lung allograft dysfunction (CLAD) is the most important immune-mediated complication of lung transplantation, representing the leading cause of late death among lung recipients. CLAD is diagnosed upon a sustained decline in lung function measures, yet the clinical course after diagnosis is highly variable. Despite the burden of CLAD, there remain no approved therapies. In part, this relates to difficulties in the design of CLAD trials due to a paucity of precise, generalizable information on CLAD progression under standard-of-care (SOC) conditions and lack of established surrogate endpoints for use in CLAD treatment studies. An additional barrier to advancing CLAD trials is the large number of participants needed and ethical concerns regarding enrollment in placebo-controlled studies for a progressive disease with substantial mortality risk. To address these gaps this proposal will create the largest multicenter cohort of lung recipients with CLAD available in the published literature and perform analyses that will inform the rational design of future CLAD trials. Specifically, to be responsive to this Notice of Special Interest (NOSI) this proposal will integrate longitudinal clinical data collected on adult lung recipients through a prior NIAID-funded Clinical Trials in Organ Transplantation (CTOT) study, CTOT-20, and through the NHLBI-funded Lung Transplant Outcomes Group (LTOG) study. Using these rich data, Aim 1 will precisely define the natural history of lung function progression after CLAD and employ latent class analyses of longitudinal lung function measures to discover subphenotypes of CLAD progression, incorporating other clinical variables that may influence disease behavior into the analytical framework. Plausible surrogate lung function endpoints for CLAD trials will be identified by assessing the association between each identified class and graft loss/death using landmarked Cox proportional hazards models. Aim 2 will evaluate the practical use of these real-world lung transplant datasets as a source of external SOC controls to complement future CLAD trials. Key eligibility criteria and endpoints assimilated from ongoing CLAD trials will be applied to evaluate how many patients from CTOT-20 or LTOG could potentially be included as real-world external controls for each trial. Simulations will then be conducted to assess operating characteristics when the external controls are used in hypothetical CLAD trials. This research is innovative because it represents a substantive departure from the existing state of CLAD investigation, namely in the application of modern statistical methods, such as latent class methods, to a large multicenter CLAD dataset and in the examination of innovative concepts, such as the use of external controls from historical data, a model of increasing interest to the FDA and industry. This research is significant because it will establish the natural history of CLAD and resolve the heterogeneity in CLAD progression in a multicenter real-world cohort; informing the design of future interventional CLAD trials in a way that is generalizable across center practices. Thus, aligned with the goals of this NOSI, these data are expected to improve the success of future CLAD trials.