Impact of tumor attachment to the pleura measured by a pretreatment CT image on outcome of stage I NSCLC treated with stereotactic body radiotherapy

Background: Pleural invasion status is known to be a predictor of survival after pulmonary resection for non-small cell lung cancer. Our goal was to determine whether the length of tumor attachment to the pleura on a pretreatment CT image has prognostic value as an alternative to pleural invasion status for stage I non-small cell lung cancer treated with stereotactic body radiotherapy (SBRT). Methods: A total of 90 tumors in 87 patients (males: 68, females: 19) who received SBRT between March 2005 and September 2011 in our institution were reviewed. The median age of the patients was 78 years (range, 48-90 years). The median tumor diameter was 2.2 cm (range, 0.9-4.2 cm). The prescribed dose was typically 48 Gy in 4 fractions, 60 Gy in 8 fractions or 60 Gy in 15 fractions to the isocenter with 6 MV X-ray using 4 non-coplanar and 3 coplanar static beams. The lengths of attachment were measured using pretreatment CT images at the lung window. Cumulative incidence rates were calculated using Kaplan-Meier curves, and univariate and multivariate analyses for in-field tumor control, locoregional control (LRC), freedom from distant metastasis and freedom from progression (FFP) were performed using a Cox proportional hazards model. Results: Of the 90 tumors, 42 tumors were attached to the pleura (median, 14.7 mm; range, 4.3-36.0 mm), 21 tumors had pleural indentation and 27 tumors had no attachment. The median follow-up period for survivors was 46.1 months. The 3-year in-field control, LRC, FFP and overall survival rates were 91.2%, 75.3%, 63.8% and 68.6%, respectively. SBRT dose and tumor diameter were independently significant predictors of in-field control (p = 0.02 and p = 0.04, respectively). Broad attachment to the pleura, the length being more than 14.7 mm, was a negative independent predictor of LRC and FFP (p = 0.02 and p = 0.01, respectively). Conclusions: Pleural attachment status on a pretreatment CT image might be an important predictor of LRC and FFP.