Multivariate Analysis Of Prognostic Factors In The Patients With Stage I Non-Small Cell Lung Cancer Treated With Hypofractionated 3-Dimensional Noncoplanar Conformal Radiation Therapy

Stereotactic body radiation therapy has become one of the standard treatments in stage I non-small cell lung cancer. However, there exists the problem of reoxygenation for large tumors and BED for serial organs locating near the central lung. Therefore, we have been treating especially these cases by decreasing the fraction dose while increasing overall treatment time and total dose (so-called hypofractionated 3-dimensional noncoplanar conformal radiation therapy). So far there has not been any difference between T1 and T2 tumors, and between peripheral and central tumors. To clarify the prognostic factors of this treatment method, we carried out this investigation using multivariate analysis (MVA). Eligibility criteria were as follows: maximum tumor diameter not greater than 5cm, PS between 0 and 2, and no limitation regarding age and pulmonary function. Radiation therapy was given with 6MV photon beam by fixed 10 non-coplanar conformal beams to a total dose of 75Gy in 25 fractions in 5 weeks. Irradiation was aiming at the ITV with proper margins. No ENI was given. Between Jan. 2002 and Jan. 2011, 109 eligible cases were treated. Age ranged from 53 to 93 (median 78). The male/female ratio was 79/30. There were 100 PS 1 and 9 PS 2 cases. There were 22 operable cases, 31 borderline operable cases (surgeons recommended RT), and 56 inoperable cases. There were 63 T1 tumors and 46 T2. Forty-six cases were central tumors and the other 63 were peripheral tumors. Seventy tumors were adenocarcinoma, 23 tumors were SCC, and 16 others. Regarding tumor markers, pretreatment CEA value (preCEA) was elevated (>5ng/ml) in 36 cases. Using these 8 parameters, MVA for overall survival (OS) and local control (LC) was performed by Cox’s Proportional Hazard Model. Median follow-up period was 67 months. Five-year LC and OS rates were 84% and 51%, respectively. As for LC, MVA revealed that histology (p=0.0279) was prognostic and PS (p=0.0541) and preCEA (p=0.0560) had a tendency. As for OS, MVA revealed that gender (p=0.0081) and preCEA (p=0.0189) were prognostic and operability (p=0.0520) and histology (p=0.0913) had a tendency. On the other hand, age, T-stage or tumor location was not prognostic regarding neither LC nor OS. The overall results of our method were promising considering the status of the patients. Regarding LC, adenocarcinomas were better controlled compared with other histologies and patients with good PS and tumors with normal preCEA tended to be better controlled. Regarding OS, female patients, patients with normal preCEA survived better than their counterpart, and operable cases and adenocarcinoma cases tended to survive better than their counterpart, respectively. On the other hand, unlike other reported series, T2 stage and central tumors did not carry worse prognoses with this treatment method, and might be recommended for this treatment.