Truncation of the PTV for Minimization of Risk of Chest Wall Toxicity in Patients Receiving SBRT for Stage I NSCLC

Abstract

Chest wall toxicity, including chest wall pain and rib fracture, is a known potential complication of stereotactic body radiation therapy (SBRT) for non-small cell lung cancer (NSCLC), impacting 10 – 40 % of patients. Smaller tumor-to-chest wall distance has been identified as a risk factor for chest wall pain after SBRT. While the standard planning target volume (PTV) is typically a uniform expansion on the gross/ internal target volume (GTV/ITV), for selected patients with tumors within 5 mm of the chest wall, we plan and deliver SBRT using a truncated PTV along the chest wall. The aims of this study were to identify outcomes and toxicities associated with use of a truncated PTV along the chest wall. We retrospectively analyzed the records of patients with stage I (T1-2aN0M0) NSCLC who received SBRT from 6/2009 to 7/2016. Patients were immobilized with a custom vacuum bag. 4-dimensional CT (4DCT) was performed. A uniform 5 mm expansion of the ITV was generated for the PTV. When this expansion overlapped with the chest wall, the region of overlap was removed from the PTV. SBRT was delivered using 3D conformal radiotherapy, intensity-modulated radiotherapy, or volumetric arc therapy on a treatment couch with six degrees of freedom with cone-beam CT (CBCT) guidance. Respiratory gating or abdominal compression were used in selected cases. For each fraction, typically at least 2 CBCTs were obtained to assess for intrafraction motion. Descriptive statistics were used to analyze outcomes. The study was approved by the IRB. Median follow-up was 36.8 months. 276 tumors were treated in 240 patients. 241 tumors were peripheral tumors and ITVs for 127 tumors (46.0%) were located within 5 mm of the chest wall. Of these, 67 tumors (52.8%) directly abutted the chest wall. Median ITV was 6.5 cc (range 0.6 – 80 cc) and median PTV was 19.5 cc (range 4.2 – 126.0 cc). Median total dose to the ITV was 48 Gy (range 48 – 60 Gy) in 4 (range 3 – 5) fractions. For the chest wall, the median V30 was 10.5 cc (range 2 – 38.8 cc), median V40 was 1.4 cc (range 0 – 23.2 cc), and median Dmax was 50Gy (range 43.8 – 60 Gy). For tumors directly abutting the chest wall, the chest wall median V30 was 11.6 cc (2.1 – 38.8 cc), median V40 was 2.9 cc (0.1 – 23.2 cc) and median Dmax was 51 Gy (45.8 – 60 Gy). The rate of chest wall pain attributable to SBRT was 1.7%. The rate of rib fracture attributable to SBRT was 4.1%; of those, only 2/10 patients reported pain. There were 15 local recurrences, and 3-year local control was 92.7%. Truncation of the PTV to avoid overlap with the chest wall in peripheral NSCLC was not associated with increased risk of local failure. There were minimal rates of chest wall toxicity despite a high percentage of tumors directly abutting the chest wall. Rigid immobilization and use of multiple CBCTs for assessment of intrafraction motion are critical when using a truncated PTV. In carefully selected patients, this is an excellent option for minimizing chest wall toxicity without compromising local control.