Overdiagnosis in chest radiographic screening for lung carcinoma: frequency

Cancer. 2003 Mar 1;97(5):1271-5. doi: 10.1002/cncr.11185.


Background: The pattern of results in the Mayo Lung Project (MLP), which is the basis for the prevailing recommendations against radiographic screening for lung carcinoma, has led to the assertion that up to 50% of the diagnosed cases of early-stage disease in that trial may have represented overdiagnosed, indolent cases. This finding suggests the possibility of such a high frequency of overdiagnosis in chest radiographic lung carcinoma screening in general. In the current study, the authors analyzed data from the MLP and its counterpart study at Memorial Sloan-Kettering Cancer Center (MSK) to estimate the frequency of overdiagnosis in these studies.

Methods: For the cases diagnosed as Stage I in the MLP and the MSK studies, the doubling times of tumor volumes were calculated. The calculations were based on size measurements recorded by the original investigators from chest radiographs taken during the course of each study.

Results: The median doubling times were 101 days in the MLP and 144 days in the MSK, times that are somewhat shorter than those reported in published series of adenocarcinoma cases diagnosed outside screening, and only 5% had doubling times exceeding 400 days.

Conclusions: The hypothesis that early-stage lung tumors diagnosed on chest radiography during lung carcinoma screening may frequently be overdiagnosed, indolent cases needs to be rejected.

MeSH terms

  • Adenocarcinoma / diagnostic imaging*
  • Adenocarcinoma / pathology
  • False Positive Reactions
  • Humans
  • Lung Neoplasms / diagnostic imaging*
  • Lung Neoplasms / pathology
  • Mass Screening
  • Neoplasm Staging
  • Radiography
  • Randomized Controlled Trials as Topic
  • Sputum