Research Lacking to Improve Drug-Induced Lung Disease Diagnosis

A descriptive review of 150-plus papers on drug-induced interstitial lung disease shows there is an absence of high-quality data on the disease.

Nazia Chaudhuri, PhD

A descriptive review of 156 full-text papers on drug-induced interstitial lung disease (DIILD) shows that there is an absence of high-quality data on the disease, causing great difficulty in assessing its impact on patients.

This investigation sprang from a project begun by the Translational Imaging in Drug Safety Initiative (TRISTAN) consortium—a team of researchers, academics and physicists looking into the development of imaging biomarkers to investigate lung and liver toxicity with the aim of earlier detection. During the review process, researchers found that very little was known about the incidence of DIILD; why it happens, how it happens, long-term impacts, and which drugs cause it.

According to information compiled on the Pneumotox website, there are over 1300 drugs, procedures, and substances that can cause respiratory problems. Currently, more than 350 pharmaceuticals have been linked to DIILD, but this often comes to light late in drug development or after launch.

“DIILD does occur during clinical trials, however it is only in a real world setting, when drugs are given to patients, often who are older or with other cormorbities and drug combinations, where we truly see the impact of DIILD regarding incidence,” said Nazia Chaudhuri, PhD, a consultant physician, with a special interest in interstitial lung disease, at the University Hospital of South Manchester and Honorary Lecturer at the University of Manchester.

The American Thoracic Society and European Respiratory Society recognize DIILD as a subtype of diffuse parenchymal lung diseases. But because clinical, pathological, and radiological features are seldom specific, it is hard to differentiate DIILD from other interstitial lung conditions.

Texts for review were taken from Medline, Embase, and the Cochrane Register of Controlled Trials during May of 2017 using the US National Library of Medicine’s Medical Subject Headings (MESH) browser.

Results showed that 78% of studies were low quality and 90% had a high risk of bias, mostly due to poor study design. There was also no clear, standardized definition of DIILD. Some used physician-reported diagnosis, others, radiological evidence.

Researchers identified six studies that most completely reported DIILD rates for specific drug types. The most common drugs linked to DIILD were cancer drugs (23-51%), anti-rheumatic drugs (6-72%), antibiotics (6-26%), non-steroidal anti-inflammatory medications (0-23%), psychiatric medications (0-9%), and anti-arrhythmic medications (0-9%).

Researchers also relied on case study reports or post-marketing studies, which can be sporadic, geographically biased, and heavily dependent on the drug studied.

Moving forward, researchers outline three proposals for further evaluation:

  • Formulating a standardized case definition for DIILD to utilize in clinical trials and observational studies.
  • Finding biomarkers for DIILD and distinguishing it from other causes of ILD.
  • Better evidence to help improve management of DIILD.

“We need to have a better understanding of the incidence of DIILD and how to detect it so that we can intervene before its gets worse and has a detrimental impact on patients,” Chaudhuri said.

The review, “Drug-Induced Interstitial Lung Disease: A Systematic Review,” was published online in the Journal of Clinical Medicine.