CyberKnife radiosurgery for inoperable stage IA non-small cell lung cancer: 18F-fluorodeoxyglucose positron emission tomography/computed tomography serial tumor response assessment

Abstract

Objective: To report serial 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET)/computed tomography (CT) tumor response following CyberKnife radiosurgery for stage IA non-small cell lung cancer (NSCLC).

Methods: Patients with biopsy-proven inoperable stage IA NSCLC were enrolled into this IRB-approved study. Targeting was based on 3-5 gold fiducial markers implanted in or near tumors. Gross tumor volumes (GTVs) were contoured using lung windows; margins were expanded by 5 mm to establish the planning treatment volumes (PTVs). Doses ranged from 42-60 Gy in 3 equal fractions. 18F-FDG PET/CT was performed prior to and at 3-6-month, 9-15 months and 18-24 months following treatment. The tumor maximum standardized uptake value (SUV(max)) was recorded for each time point.

Results: Twenty patients with an average maximum tumor diameter of 2.2 cm were treated over a 3-year period. A mean dose of 51 Gy was delivered to the PTV in 3 to 11 days (mean, 7 days). The 30-Gy isodose contour extended an average of 2 cm from the GTV. At a median follow-up of 43 months, the 2-year Kaplan-Meier overall survival estimate was 90% and the local control estimate was 95%. Mean tumor SUV(max) before treatment was 6.2 (range, 2.0 to 10.7). During early follow-up the mean tumor SUV(max) remained at 2.3 (range, 1.0 to 5.7), despite transient elevations in individual tumor SUV(max) levels attributed to peritumoral radiation-induced pneumonitis visible on CT imaging. At 18-24 months the mean tumor SUV(max) for controlled tumors was 2.0, with a narrow range of values (range, 1.5 to 2.8). A single local failure was confirmed at 24 months in a patient with an elevated tumor SUV(max) of 8.4.

Conclusion: Local control and survival following CyberKnife radiosurgery for stage IA NSCLC is exceptional. Early transient increases in tumor SUV(max) are likely related to radiation-induced pneumonitis. Tumor SUV(max) values return to background levels at 18-24 months, enhancing 18F-FDG PET/CT detection of local failure. The value of 18F-FDG PET/CT imaging for surveillance following lung SBRT deserves further study.

Figure 1

Kaplan-Meier plot of overall survival.

Figure 2

Change in mean tumor SUV_max.

Figure 3

Individual patient changes in tumor SUV_max.

Figure 4

Right upper lobe clinical stage IA NSCLC treatment planning PET/CT with a tumor SUV_max of 10.5 (A), planned radiation dose distribution (B: the planning treatment volume receiving 45 Gy shown in red and the 30 Gy isodose line in blue), and PET/CT at 6, 12, and 18 months post-treatment (C, D and E) show an initial decrease in tumor SUV_max to 1.5 followed by a transient radiation induced increase (tumor SUV_max = 4.0) which resolves by 18 months (tumor SUV_max = 2.5).

Figure 5

Right upper lobe clinical stage IA NSCLC treatment planning PET/CT with a tumor SUV_max of 8.7 (A), planned radiation dose distribution (B: the planning treatment volume receiving 45 Gy in red and the 30 Gy isodose line in blue), and PET/CT at 12, and 24 months post-treatment (C and D) show an initial decrease in SUV_max to 2.3 followed by local recurrence (SUV_max = 8.4).

Figure 6

Recurrent tumor (bold arrow) infiltrating radiation-induced lung fibrosis (dashed arrow).

Figure 7

Kaplan-Meier plot of local control.