Predictors of Pathologic Tumor Invasion and Prognosis for Ground Glass Opacity Featured Lung Adenocarcinoma
Ting Ye,Lin Deng,Jiaqing Xiang,Yawei Zhang,Hong Hu,Yihua Sun,Yuan Li,Lei Shen,Shengping Wang,Li Xie,Haiquan Chen,Quiana L. Kern,Chadrick E. Denlinger
DOI: https://doi.org/10.1016/j.athoracsur.2018.06.058
IF: 5.113
2018-01-01
The Annals of Thoracic Surgery
Abstract:Background. We make surgical strategies for ground glass opacity (GGO) nodules currently based on thinsection (TS) computed tomography (CT) findings. Whether radiologic measurements could precisely predict tumor invasion and prognosis of GGO-featured lung adenocarcinoma is uncertain. Methods. We retrospectively evaluated medical records of patients with radiologic GGO nodules undergoing a surgical procedure at Fudan University Shanghai Cancer Center. The study endpoints were the predictive value and prognostic significance of radiologic measurements (consolidation-to-tumor ratio value, consolidation size, and tumor size) for pathologic lung adenocarcinoma. Results. In this study 736 patients and 841 GGO nodules were included. Five-year lung cancer-specific regression-free survival (LCS-RFS) rate was 95.76% (95% confidence interval [CI], 93.01% to 97.44%). The 5-year LCS overall survival (OS) rate was 98.99% (95% CI, 97.69% to 99.57%). Multivariable analysis showed that tumor invasion (invasive adenocarcinoma [IAD] vs adenocarcinoma in situ [AIS]/minimally invasive adenocarcinoma [MIA], p = 0.020) was the only independent predictor for 5-year LCS-RFS. IAD (hazard ratio, 15.98; 95% CI, 1.55 to 164.35) was correlated with a higher risk of recurrence. Kaplan-Meier analysis showed that only tumor invasion status (IAD vs AIS/MIA, p = 0.003) could predict 5-year lung cancer-specific overall survival (LCS-OS), and IAD had a worse LCS-OS than AIS and MIA. A part-solid component (odds ratio [OR], 9.09; 95% CI, 2.71 to 30.47; p = 0.000), large consolidation size (OR, 3.11; 95% CI, 1.03 to 9.40; p = 0.045), and large tumor size (OR, 5.48; 95% CI, 2.68 to 11.19; p = 0.000) were associated with pathologic IAD. For IAD <= 20 mm, segmentectomy and lobectomy had better 5-year LCS-RFS than wedge resection, although the difference was statistically insignificant (p = 0.367). The three types of surgeries provided the similar 5-year LCS-OS (p = 0.834). Conclusions. Radiologic measurements could not precisely predict tumor invasion and prognosis. Making treatment strategies solely according to TS-CT findings for GGO tumor is inappropriate. (C) 2018 by The Society of Thoracic Surgeons