Development and validation of a prediction model for the indication of inhaled corticosteroid (ICS) in patients with chronic obstructive pulmonary disease (COPD)

Lead Investigator: Yeon-Mok Oh, Asan Medical Center
Title of Proposal Research: Development and validation of a prediction model for the indication of inhaled corticosteroid (ICS) in patients with chronic obstructive pulmonary disease (COPD)
Vivli Data Request: 6712
Funding Source: None
Potential Conflicts of Interest: None

Summary of the Proposed Research:

Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide. COPD is well known as a clinical syndrome characterized by chronic respiratory symptoms, structural pulmonary abnormalities, lung function impairment, or any combination of these. Inhaled drugs, including long-acting bronchodilators and steroids, are mainly used for COPD pharmacotherapy. Among them, the role of inhaled corticosteroid (ICS) for COPD patients are less known than bronchodilators. Pneumonia, which can be detrimental for COPD patients, is the well-known side effect for ICS. Therefore, when prescribing ICS to COPD patients, its positive effects and side effects must be fully considered. In this study, we intend to create a model that predicts COPD patients who can expect a positive effect when using ICS. Through this study, we expect to improve the quality of life and provide better therapeutic effects for patients requiring ICS in COPD patients.

Statistical Analysis Plan:

Risk Score Development: Broad aspects of potential risk factors, including history of drugs and pulmonary function, will be examined by Cox regression analyses using the data from IMPACT study. To identify heterogeneity of treatment effects, interaction effects of risk factors with ICS use will be assessed. Depending on the cox analysis results, risk scores that varies by ICS use will be developed. Weights for the risk scores will be derived from β-coefficients of the final models. For easier interpretation and implementation, we will also explore possible categorization of continuous predictors, and a point scoring system will be considered. For example, points may consist of integers derived from β-coefficients divided by the lowest β-coefficient of all predictors chosen and rounded to the nearest integer.

External Validation: Prediction accuracy of the developed risk scores will be evaluated through the data from TRIBUTE study. To assess model discrimination, c-index will be examined. To assess model calibration, decile plots or calibration curves (plots of predicted versus observed response) will be inspected.

Requested Studies:

A Phase III, 52 Week, Randomized, Double-blind, 3-arm Parallel Group Study, Comparing the Efficacy, Safety and Tolerability of the Fixed Dose Triple Combination FF/UMEC/VI With the Fixed Dose Dual Combinations of FF/VI and UMEC/VI, All Administered Once-daily in the Morning Via a Dry Powder Inhaler in Subjects With Chronic Obstructive Pulmonary Disease
Data Contributor: GlaxoSmithKline
Study ID: NCT02164513
Sponsor ID: CTT116855

Public Disclosures:

  1. Lee JH, Kim S, Oh YM. A Prediction Scoring Model for the Effect of Withdrawal or Addition of Inhaled Corticosteroids in Patients with Chronic Obstructive Pulmonary Disease. Int J Chron Obstruct Pulmon Dis. 2023;18:113-127. doi : 10.2147/COPD.S389502
  2. Jang Ho Lee, Sehee Kim, Yeon-Mok Oh. Development of a scoring model for the prediction of the effects of discontinuation or addition of inhaled corticosteroid in chronic obstructive pulmonary disease patients. Volume 28, Issue S1 Supplement: 26th Congress of the Asian Pacific Society of Respirology Above and Beyond, 17–20 November 2022, Seoul, Korea, February 2023, P45-46 AO07-3. doi: 10.1111/resp.14433#14433-sec-0269