Computed Tomography Study Assesses Model for Predicting Recurrence of Non-Small Cell Lung Cancer

Can an emerging predictive model, which incorporates preoperative computed tomography (CT) findings and clinical data, enhance stratification for patients at risk for post-op recurrence of non-small cell lung cancer (NSCLC)?

In a retrospective study, recently published in Clinical Imaging, researchers developed and trained a predictive model, which incorporated patient age, tumor markers, pleural effusion, and consolidation/tumor ratio (CTR), in 438 patients who had surgically resected NSCLC. The model was subsequently evaluated in a testing cohort (188 patients) and an external validation cohort (69 patients)

The study authors found that the predictive model had an 86.7 percent area under the curve (AUC), a 75.5 percent accuracy rate, 94.4 percent sensitivity and 73.5 percent specificity in the testing cohort. External validation testing revealed an 85.2 percent AUC, 79.7 percent accuracy, 83.3 percent sensitivity and 78.9 percent specificity, according to the researchers.

“The predictive information about the risk of recurrence of NSCLC provided by this model may allow patients to develop more individualized treatment plans and provide more accurate prognostic assessments for patients,” wrote lead study author Xinjie Yu, M.D., who is affiliated with the Department of Radiology at the Tongde Hospital of Zhejiang Province in Hangzhou, China, and colleagues.

While noting studies that have demonstrated the prognostic capability of positron emission tomography/computed tomography (PET/CT) in assessing risks for recurrence and metastasis in patients with lung cancer, the study authors maintained that CT offers a more practical imaging option in this patient population.

“CT is the tool most widely used for detecting and evaluating lung cancerand offers advantages such as accurate three-dimensional visualization, clear display of internal details, and comprehensive assessment of surrounding tissues. Compared with PET–CT scans, CT scans are more cost-effective with a shorter scanning time while providing higher spatial resolution,” posited Yu and colleagues.

Emphasizing that pleural effusion was one of the key parameters utilized in the development of the predictive model and is easy to detect on CT, the researchers noted that malignant pleural effusion is commonly associated with metastatic disease and poor prognosis.

“Malignant pleural effusion affects up to 15% of patients with cancer and is most common in those with lung cancer, breast cancer, lymphoma, gynecological malignancy, and malignant mesothelioma,” noted Yu and colleagues.

(Editor’s note: For related content, see “What Emerging CT Research Reveals About Obesity and Post-Op Survival for Non-Small Cell Lung Cancer,” “CT Study Reveals Key Indicators for Angiolymphatic Invasion in Non-Small Cell Lung Cancer” and “CT Study Links Better Five-Year Prognosis with Minor Ground Glass Opacity Component in NSCLC Lung Nodules.”)

In regard to study limitations, the authors acknowledged a small cohort size (69 patients) for external validation testing and possible patient selection bias due to the retrospective study design.