Long-term toxicity in patients receiving radiotherapy for ultracentral stage i non-small cell lung cancer - a secondary analysis of the lustre randomized trial

Hypofractionated and stereotactic body radiotherapy (SBRT) are increasingly used in the treatment of centrally located, early-stage non-small cell lung cancer (NSCLC), though there are concerns of increased morbidity and mortality in patients with ultracentral tumors (UC). We report on the long-term toxicity of patients with UC lung cancer treated on a prospective randomized clinical trial of SBRT versus conventionally hypofractionated radiotherapy (CRT) for stage I NSCLC (NCT01968941).Patients with UC tumors, defined as those where the planning target volume directly overlaps with the proximal bronchial tree (PBT), were identified from the larger cohort of patients treated on the trial. These patients received either SBRT with 60 Gy in 8 fractions or CRT with 60 Gy in 15 fractions. The primary endpoint of this secondary analysis was development of any grade 3 or higher toxicity defined using CTCAE version 3.0. Secondary endpoints included local control, as well as dosimetric analysis of the PBT, using EQD2 with α/β ratio of 3 to assess the relationship between dose to the PBT and toxicity.Twenty-nine patients were identified with UC tumors; 21 received SBRT and 8 received CRT. Median age was 72 years (range 55-88 years) and 59% were female. Median FEV1 was 1.46L (range 0.64-2.37L). Patients had either T1 (59%) or T2 (41%) lesions, with median tumor size 2.5cm (range 1.1-4.9cm). Most patients had histologically confirmed disease (squamous cell, n = 10; adenocarcinoma, n = 8; radiographically suspicious, n = 11). The median follow-up was 2.9 years (range 0.7-5.2 years). The 3-year local control rate of all patients was 88.3% (95% confidence interval: 75.7-100%). There were 3 patients with late (>3 months) grade 3 toxicity (bronchial stricture, chest pain, and atelectasis) and 1 patient with late grade 5 toxicity (bleeding/hemorrhage), all treated in the SBRT arm. Median EQD2 dose to PBT in patients with grade ≥3 late toxicity compared to the rest of the cohort was: Dmax, 132 vs 129 Gy; D0.1cc, 129 vs 119 Gy; D1cc, 124 vs 80 Gy; and D5cc, 83 vs 41 Gy. Median EQD2 volumetric doses in grade ≥3 patients (compared to the rest) to PBT were: V65 Gy, 9.7 vs 2.2cc; V80 Gy, 7.9 vs 1.1cc; V90 Gy, 6.2 vs 0.4cc; and V100 Gy, 4.8 vs 0.3cc. The single patient with grade 5 toxicity had the highest D5cc (116 Gy) and V100 Gy (7cc) among all patients.Stereotactic radiation with 60 Gy in 8 fractions for UC lung cancer provides good local control but carries an approximately 15-20% rate of late grade ≥3 toxicity. There appears to be a dosimetric association between toxicity and dose to the PBT. It may be more important to minimize volumetric PBT dose rather than maximum point dose to reduce risk of severe late toxicity.