Effectiveness of CT Histogram Analysis to Differentiate Lung Metastases From Second Primary Lung Cancer to Decrease Need for Lung Biopsy.

Abstract

Objective: Differentiating lung metastasis from second primary lung cancer is crucial for determining the appropriate treatment strategy. Lung biopsy, the gold standard for diagnosis, is an invasive procedure. This study aimed to evaluate the potential of CT histogram analysis as a noninvasive method for differentiating these 2 conditions in solitary pulmonary nodules.

Methods: A retrospective analysis was conducted on CT images of patients with solitary pulmonary nodule, confirmed to be either lung metastasis or second primary lung cancer histopathologically. Histogram analysis features of the lesion and perilesional area were extracted from the CT images and subjected to statistical analysis to identify significant differences between the 2 groups. The performance of histogram analysis was assessed using sensitivity, specificity, and area under the ROC curve.

Results: The data of 26 (46%) patients whose lung biopsy pathology was determined as second primary lung cancer and 30 (54%) patients defined as lung metastasis were investigated. The second primary lung cancer's mean pathologic tumor diameter was statistically higher than the lung metastasis [25.3 (5.7) mm, 18.3(5.6) mm; P=0.003]. The mean skewness (P=0.020) and entropy (P=0.018) values in the second primary lung cancer were statistically significantly lower in the lesion area. There was a statistically significant difference in the mean measurement of SD (P=0.001), skewness (P<0.001), kurtosis (P<0.001), and entropy (P<0.001) values between the 2 groups in the perilesional area.

Conclusion: CT histogram analysis shows promise as a noninvasive method for differentiating lung metastasis from second primary lung cancer in solitary pulmonary nodules.