Comparison Between Stereotactic and Conventional Radiotherapy for Solitary Lung Tumor After Resection of Lung Cancer

Patients and Methods

Eligibility criteria. This retrospective analysis was approved by our institutional review board with approval number H28-111. Patients were considered eligible for this study on the basis of the following criteria: i) patients had a solitary lung tumor that was a postoperative local recurrence (PLR), pulmonary metastasis (PM), or multiple primary lung cancer (MPLC) following the complete resection of an NSCLC; ii) RT alone was performed for the solitary lung tumor between 2006 and 2015 in our institution; and iii) none of the patients had regional or extra-pulmonary distant metastasis at the time of RT. Patients treated with concurrent chemoradiotherapy were excluded.

Subgroup criteria of PLR, PM, or MPLC. PLR was defined as a lesion adjacent to a staple line or a bronchial stump following the complete resection of an NSCLC. As for the lesions excluding PLR, PM and MPLC were differentiated, according to the criteria by Martini et al. (13). The criteria for MPLC are as follows: I) different histology between an MPLC and a completely resected NSCLC or II) same histology between an MPLC and a completely resected NSCLC, with i) more than 2 years tumor-free interval, ii) within 2 years of a tumor-free interval in a different lobe or lung, or iii) with origin from carcinoma in situ.

Treatment. A vacuum cushion was used to fix the body of the patients who were treated with SBRT. The planning computed tomography (CT) scan under free-breathing was performed using a long-time scan (3 seconds/scan) to calculate internal target volume (ITV). CT volume data were then transferred to a 3-dimensional radiation treatment planning system. A physician delineated the target volume on the planning CT. Spicula formation and pleural indentation were included within the ITV. Planning target volume (PTV) margin from the ITV was 5-15 mm at the physicians' discretion with reference to the respiratory motion estimated via the use of an X-ray simulator.

Eight ports including 4 coplanar and 4 noncoplanar beams were used for SBRT. Three or 4 coplanar beams were used for CFRT. A heterogeneous dose calculation algorithm was used. RT was delivered using 4 or 10 MV and 6 MV photon of a linear accelerator until June 2008 and from July 2008, respectively.

SBRT using 48 Gy in 4 fractions over 1-2 weeks at the isocenter was administered to the patients with a peripheral solitary lung tumor (SBRT group). CFRT using 66-70 Gy in 33-35 fractions over 6.6-7 weeks at the isocenter was administered to the patients with a central solitary lung tumor adjacent to the pulmonary hilum or the mediastinum, to avoid a risk of severe hilar or mediastinal toxicity with SBRT (CFRT group).

Follow-up. Patients were followed up 1 and 3 months following the completion of RT. Subsequently, patients were observed every 3 months. During each follow-up visit a chest X-ray was performed. CT or fluorodeoxyglucose (FDG) positron emission tomography (PET)/CT was performed every 6-12 months to evaluate recurrence. Tumor markers were assessed at the physicians' discretion.

Local recurrence was defined as tumor progression or relapse in the irradiated field. Recurrence which occurred within the lung but appeared outside the post-irradiated change was scored as distant recurrence within the lung. Regional recurrence was defined as the regional lymph node recurrence. Any other recurrence was defined as distant recurrence.

Statistics. OS was defined as the time from the completion of RT to death due to any cause. Local control (LC) was defined as the time from the completion of RT to local recurrence. The Kaplan–Meier method was used to calculate the OS and LC rates. Regarding the OS and LC rates, the log-rank test was used in the analysis. Common Terminology Criteria for Adverse Events, version 4.0 (14), were used to assess the toxicities. Fisher's exact test was used in the analysis of the patient and tumor characteristics and toxicities. A p-value of <0.05 was considered to indicate statistical significance. Statistical analyses were performed using the JMP software, version 11 (SAS Institute, Cary, USA).