Clinical predictor of pre- or minimally invasive pulmonary adenocarcinoma: possibility of sub-classification of clinical T1a

Abstract

Objectives: A new pathological classification for pre- and minimally invasive adenocarcinoma has been established, with distinction prior to surgery crucial because of the extremely good prognosis.

Methods: Of 412 patients who underwent surgery for lung cancer from 2008 to 2011, 110 classified as c-stage I had each of the following four parameters assessed for predictive power for pre- or minimally invasive adenocarcinoma and relapse-free survival (RFS): (i) whole tumour size (WS) shown by computed tomography (CT) , (ii) size of the solid (SS) component in CT findings, (iii) maximum standard uptake value in fluorodeoxyglucose positron emission tomography (FDG-PET)/CT scan images (SUVmax) and (iv) serum level of carcinoembryonic antigen.

Results: For prediction of pre- or minimally invasive adenocarcinoma, the area under the receiver-operating curve was >0.7 for all the four parameters, while only SS was found to be an independent factor in multivariate logistic regression analysis. In Cox proportional hazard model analysis, SS and SUVmax were statistically significant, and SS was exclusively independent in multivariate analysis. Differences in RFS between T1a and T1b were more pronounced when using SS compared with WS. In the sub-classification of T1a, we used a breakpoint of 1.0 cm in SS (T1a-α and T1a-β), which resulted in a 2-year RFS rate of 1.00 for T1a-α (n=21), 0.89 for T1a-β (n=27) and 0.68 for T1b (n=26) (P=0.002 between T1a-β and T1b).

Conclusions: The SS parameter was useful to distinguish pre- and minimally invasive adenocarcinoma from other types of lung cancer, and set a T1a sub-classification.

Keywords: Carcinoembryonic antigen; Computed tomography; Invasive adenocarcinoma; Non-small-cell lung cancer; SUVmax; Surgery.