Overdiagnosis in lung cancer screening

Area of Science:

• Oncology
• Radiology
• Epidemiology

Background:

• Overdiagnosis in lung cancer screening is a significant concern, with numerous publications highlighting potential harms.
• The current definition of overdiagnosis relies on epidemiological factors (cause of death) rather than clinical presentation, leading to ambiguity.
• This epidemiologically-based definition is unusual and contributes to confusion regarding the frequency of overdiagnosis.

Purpose of the Study:

• To clarify the definition and frequency of overdiagnosis in lung cancer screening.
• To address the confusion arising from the current epidemiologic definition of overdiagnosis.
• To propose a clinically-based definition focusing on cancer aggressiveness.

Main Methods:

• Review of existing literature and screening trial data, including the National Lung Screening Trial (NLST).
• Analysis of the limitations of the current epidemiologic definition of overdiagnosis.
• Exploration of a proposed clinical definition based on cancer aggressiveness and presentation.

Main Results:

• Contrary to earlier concerns, extended follow-up of the NLST indicates virtually no overdiagnosis in CT screening.
• Previous assumptions of high overdiagnosis rates with chest radiography and even higher with more sensitive CT screening were not supported by NLST data.
• A proposed clinical definition could identify overdiagnosis based on misclassifying indolent cancers (e.g., nonsolid nodules) as aggressive.

Conclusions:

• The current epidemiologic definition of overdiagnosis in lung cancer screening is problematic and leads to confusion.
• Evidence from the NLST suggests overdiagnosis is infrequent in CT lung cancer screening.
• A shift towards a clinical definition, focusing on cancer aggressiveness, is proposed to better identify overdiagnosis and prevent overtreatment.