Diagnostic Performance of Imaging Methods in Predicting Lung Cancer Metastases

  • 0Department of Radiology, Faculty of Medicine, Prof Dr Süleyman Yalçin City Hospital, Istanbul Medeniyet University, Istanbul, Turkey.

|

|

Summary

This summary is machine-generated.

An algorithm analyzing lung mass imaging can predict distant organ metastasis in lung cancer patients. Bone metastasis was most common, linked to specific lesion characteristics and patient factors.

Area Of Science

  • Radiology
  • Oncology
  • Medical Imaging Analysis

Background

  • Lung cancer is a leading cause of cancer-related mortality worldwide.
  • Early detection and prediction of metastasis are crucial for effective treatment planning.
  • Radiological imaging plays a key role in diagnosing and staging lung cancer.

Purpose Of The Study

  • To investigate the potential of a novel algorithm for predicting distant organ metastasis in lung cancer.
  • To evaluate the relationship between lung mass morphology, localization, and metastatic sites.
  • To identify imaging features associated with increased risk of metastasis.

Main Methods

  • Retrospective analysis of lung cancer patients diagnosed between 2016-2023.
  • Utilized an algorithm to assess lesion morphology, proximity to vital structures, and maximum standardized uptake value (SUVmax).
  • Investigated metastasis to six common sites: contralateral lung, liver, brain, adrenal glands, bone, and others.

Main Results

  • 383 patients included; 106 (43.8%) had distant metastases.
  • Bone metastasis was most frequent (42.5%), particularly in male patients.
  • Metastasis risk was higher for lesions adjacent to ribs/bronchi, involving mediastinal lymph nodes, with irregular contours, and SUVmax > 11.15.

Conclusions

  • An algorithm assessing morphological and neighborhood characteristics of lung masses can predict metastasis.
  • Radiological imaging features offer predictive value for the location and likelihood of lung cancer metastasis.
  • This approach aids in understanding and potentially forecasting metastatic patterns in lung cancer.