Neither Maximum Tumor Size nor Solid Component Size Is Prognostic in Part-Solid Lung Cancer: Impact of Tumor Size Should Be Applied Exclusively to Solid Lung Cancer

Abstract

Background: We aimed to investigate the prognostic impact of tumor size based on the consolidation status by thin-section computed tomography.

Methods: We evaluated 1,181 surgically resected clinical N0 M0 non-small cell lung carcinomas. Consolidation tumor ratio (CTR) was evaluated for all, and tumors were classified into three groups, namely pure ground-glass opacity (CTR = 0; n = 168), part-solid (0 < CTR < 1.0; n = 448), and solid (CTR = 1.0; n = 565). The impact of tumor size was assessed based on CTR using Cox proportional hazards model.

Results: Tumor size significantly differentiated the 5-year overall survival (≤20 mm; n = 638: 93.4%; 21-30 mm; n = 284: 84.2%; 31-50 mm; n = 193: 69.3%; ≥51 mm; n = 66: 43.5%; p < 0.0001). When we evaluated the impact of tumor size based on CTR, the 5-year overall survival differed significantly in patients with radiologic solid lung cancer (≤20 mm: 83.0%; 21-30 mm: 75.4%; 31-50 mm: 56.2%; ≥51 mm: 45.3%; p < 0.0001). In contrast, it did not affect the 5-year overall survival in patients with radiologic pure ground-glass opacity (100% regardless of the tumor sizes) and part-solid lung cancer (≤20 mm: 97.7%; 21-30 mm: 94.6%; 31-50 mm: 93.4%; p = 0.1028; 0 < CTR ≤ 0.5: 98.4%; 0.5 < CTR < 1.0: 95.0%; p = 0.1247). Furthermore, maximum tumor size (p = 0.6370), solid component size (p = 0.2340), and CTR (p = 0.1395) were not associated with poor overall survival in radiologic part-solid lung cancer.

Conclusions: The impact of maximum tumor size should be applied only to radiologic solid lung cancer without the ground-glass opacity component on thin-section computed tomography. On the other hand, we recommend that pure ground-glass opacity and part-solid lung cancers be described, respectively, as clinical-Tis and clinical-T1a, which are independent of maximum tumor size and solid component size on thin-section computed tomography.