Stereotactic Lung Radiotherapy (Sbrt) Associated With Improved Local Control Compared To 3-Dimensional Conformal Radiotherapy (3dcrt) For Stage I Non-Small Cell Lung Cancer (Nsclc)

To compare lung SBRT with standard fractionated 3DCRT for treatment of early-stage NSCLC in terms of local control, overall survival, and associated toxicities. One hundred thirty-five patients (136 lesions) with stage I NSCLC comprised two treatment modality groups; the 3DCRT group had 54 patients, consisting of T1N0M0: 65% and T2N0M0: 35%. The SBRT group had 81 patients (82 primary lesions) consisting of T1N0M0: 70% and T2N0M0: 30%. The median patient ages were 76y, 75y for SBRT, 3DCRT respectively; 57% were women for SBRT while 61% were women for 3DCRT (p = NS). More African Americans had SBRT (15% v 5%, p<0.01). All patients were medically inoperable. Median percentage predicted DLCO was 53% and 42% (p<0.01) for 3DCRT and SBRT, respectively. There were no differences in percentage predicted FEV1 between groups (62% SBRT v 59% 3DCRT). Although overall T-stages were similar between groups, 3DCRT patients had larger tumors with a median size of 2.7cm for SBRT v 3.6 for 3DCRT (p<0.01). The median dose for 3DCRT was 70 Gy in 35 fractions (54-79.2 Gy), and for SBRT was 48 Gy in 4 fractions (40-60 Gy). Follow-up was ≥2y in both groups. With a mean follow-up of 1.57 years, SBRT had a significantly lower rate of local recurrence (LR) at 2 years, 8%, compared to 21% with 3DCRT (p = 0.02). No statistically significant differences were identified, however, in the rates of regional recurrence (RR) 12% SBRT v 9% 3DCRT (p = NS) or distant metastasis (DM) 19% in both groups. Disease-free survival was also similar in both groups, 49% in each. Although overall survival (OS) and cause-specific survival (CSS) were somewhat higher in the SBRT group, these differences did not reach statistical significance in this patient dataset. 2y OS was 65% after SBRT v 54% after 3DCRT (p = NS); 2y CSS was 91% after SBRT and 80% after 3DCRT (p = NS). The only statistically significant difference in treatment-related toxicity between the two groups was a higher rate of ≥ grade 2 dermatitis after 3DCRT (11% v 1%, p = 0.02). Otherwise, the rates of ≥ grade 2 pneumonitis (6% SBRT v 5% 3DCRT), acute esophagitis (0% v 2%), chronic myositis (5% v 2%) and rib fracture (5% v 0%) were statistically similar. For all patients, no statistically significant predictor of LR could be identified other than 3DCRT. Only gender (Male) predicted RR (p<0.01) or DFS (p = 0.02). In this dataset, online image-guided SBRT was associated with superior local control versus standard fractionation 3DCRT for stage I NSCLC, but not a significant improvement in survival. Toxicity profiles were similar between the two groups. These findings support the continued use of hypofractionated SBRT.