(P071) SBRT Treatment of Central Chest Lesions: Experience at the University of California, San Francisco

Julian Johnson, MD, Steve Braunstein, MD, PhD, Martina Descovich, PhD, Cynthia Chuang, PhD, Alexander Gottschalk, MD, PhD, Sue S. Yom, MD, PhD; University of California, San Francisco

Introduction: Early experiences with stereotactic body radiotherapy (SBRT) for treatment of intrathoracic lesions have demonstrated excellent local control rates, upwards of 90% for primary and metastatic lesions in the lung parenchyma. However, there exists concern over normal tissue toxicity of central lung structures with such hypofractionated regimens; several single-institution series have recently reported varying rates of toxicity, and results of the Radiation Therapy Oncology Group (RTOG) 0813 trial are pending. We describe our institutional experience with SBRT treatment of central lung lesions and patient outcomes.

Materials and Methods: We conducted a retrospective review of all patients who underwent robotic SBRT to thoracic lesions at UCSF from 2004–2012 (n = 482). We identified 90 patients who were treated for central lung lesions, defined as within 2 cm of the proximal bronchial tree. We describe patient and disease characteristics as related to locoregional control on routine interval imaging, as well as toxicity, as graded by Common Terminology Criteria for Adverse Events version 4 (CTCAE v4).

Results: Of the 90 patients treated for central lung lesions, 18 patients were treated for primary lung tumors and 72 had recurrent or metastatic disease. Fifty-six percent of all patients had prior thoracic surgery, and 28% of all patients had prior thoracic radiotherapy. Median age at treatment was 65 years (range: 13–83 yr). Median Karnofsky performance score (KPS) was 80 (range: 40–90). Average lesion size was 2.7 ± 1.3 cm maximum transverse diameter (range: 0.5–6.2 cm). Average biological effective dose (BED) to the planning target volume (PTV) was 83 ± 30, over one to six fractions (median five), with dose per fraction ranging from 4–20 Gy. Median BED was 100 due to a large number of patients being treated with 50 Gy in five fractions (range: 20–180). With a median follow-up of 8 months (range: 0–71 mo, 12 mo primary lung, 12 mo living patients), 63% of patients reported no acute or late toxicity. One patient experienced acute grade 3 toxicity, consisting of radiation pneumonitis. Three patients experienced late-grade 3 toxicity, which were of respiratory or cardiac etiology. BEDs were 72, 100, and 100. Of the two patients who experienced grade 3 pulmonary toxicity, both had prior thoracotomies. None had prior RT. One patient’s tumor was within the mainstem bronchus. The patient with cardiac toxicity had a prior history of arrhythmia and coronary artery disease. The Kaplan-Meier estimate of locoregional control for all patients was 69% at 1 year.

Conclusion: We demonstrated local control of nearly half of all treated lesions associated with minimal toxicity. To increase local control, in the future we will raise the BED to treated lesions. Three percent of patients experienced greater than grade 2 treatment-related toxicity, which may have been confounded by systemic disease progression or prior surgery.