JMIR Publications

ABSTRACT

Background:

As the importance of real-world data (RWD) as a source of evidence for various drug effects has increased, clinical research utilizing the secondary use data from electronic medical record systems is gaining increasing attention. While immune checkpoint inhibitors and targeted therapies have advanced regarding lung cancer treatment, managing complications, such as interstitial lung disease (ILD), is challenging. The early detection and prevention of ILD are crucial for improving patient prognosis and quality of life; however, predictive biomarkers remain unestablished. Therefore, methods for identifying ILD risk factors and enabling early detection using RWD are required.

Objective:

This exploratory study aimed to identify the potential risk factors and prodromal symptoms of ILD using clinical data stored in a hospital information system.

Methods:

Clinical data of patients diagnosed with lung cancer between November 2011 and December 2018 were extracted from the hospital information system of the National Cancer Center Hospital in Japan. Three sets were defined: the ILD Set, defined by laboratory test results and radiological records; the ILD-GC Set, which added glucocorticoid treatment to the definition of the ILD Set; and the No ILD Set for patients without ILD. The primary endpoint was the frequency of Japanese words in the electronic medical records extracted from the POS/SOAP notes. Additionally, laboratory test results were evaluated for each time window, and baseline characteristics and anti-cancer drug use were assessed. The free text was processed using morphological analysis. Variations in descriptions were corrected using the Patient Disease Expression Dictionary or the WHO Drug Dictionary. Important terms were extracted from physician and nurse records to identify characteristic findings and symptoms prior to ILD onset.

Results:

The analysis included 674 cases (105 in the ILD Set [including 12 in the ILD-GC Set] and 569 in the No ILD Set). The baseline characteristics showed no apparent differences. Moreover, the use of anti-cancer drugs, programmed cell death protein-1 inhibitors, epidermal growth factor receptor tyrosine kinase inhibitors, anaplastic lymphoma kinase tyrosine kinase inhibitors, and cytotoxic agents were confirmed. The KL-6 and surfactant protein D levels were not abnormal immediately before ILD onset. Thirty days prior to ILD onset, the frequency of terms related to respiratory symptoms (Breathlessness, Shortness of breath, O2), pain/analgesics (Lyrica (Product name of pregabalin), Soreness, Pain-Precordial, Opioid, etc.), and appetite (Inappetence, Food-intake, Feeling-queasy, Novamin (Product name of prochlorperazine)) increased.

Conclusions:

This study used RWD from medical institutions providing acute treatment for malignant tumors to identify terms related to respiratory symptoms and pain/analgesics as risk factors for ILD onset in patients with stage IV lung cancer. The findings may contribute to the early detection of ILD. Furthermore, this study highlights RWD’s potential to generate real-world evidence applicable to drug discovery and pharmaceutical development.