SBRT has emerged as the preferred treatment approach for stage I NSCLC in patients with peripheral tumors who refuse surgery or are deemed medically inoperable. Evidence of efficacy in lung cancer has been accumulating since 1995 and mostly comes from retrospective observational series (single and multi-institution) and some prospective phase I/II clinical trials. Many of the trials are populated with patients who refuse surgery and patients who are deemed too high risk for surgery. One of the largest contributions to the SBRT literature (with >5-year follow-up) comes from Japan where investigators reported on 257 patients from 14 different hospitals and showed the importance of dose response. The local recurrence rate was significantly lower for a BED (biologic effective dose) of ≥100 Gy compared with a BED <100 Gy (8.4 versus 42.9 %, p = 0.01). Disease specific 5-year survival was 73.2 % in the total cohort where 99 patients were considered operable. The overall 5-year survival rates of medically operable and inoperable patients were 64.8 and 35.0 %, respectively [5]. The improved survival in operable patients treated with SBRT has been corroborated in other prospective series [13, 14] and likely speaks to the negative influence of medical comorbidities in high risk patients. Data supporting the use of standardized SBRT dosing in stage I NSCLC in North America was examined in strictly medically inoperable patients in a phase II multi-institution study (RTOG 0236) [7]. Fifty-five patients with T1/T2 NSCLC were treated with 54 Gy (three fractions × 18 Gy) and followed for recurrence and survival over 2 years. Four patients failed at the primary site or within the same lobe rendering a 3-year local control rate = 91 %. Combining local and regional failures, the 3-year local-regional control rate was 87 %. Disease-free survival and overall survival at 3 years were 48.3 and 55.8 %, respectively.

The lack of dose uniformity and optimal fractionation of SBRT for stage I NSCLC can be seen in Table 15.2. Many of these studies are also limited by a median follow up ≤36 months. Results from both retrospective and prospective series (Table 15.2) show 3-year OS and DFS approaching 57 and 81 %, respectively [13, 15–17]. Five year overall and cancer specific survival after SBRT treatment are mostly absent in the literature thus making comparisons to surgical series of stage I NSCLC rather limited. Although some of the series have operable patients that refuse surgery [5, 13, 17, 18], there are few data to render conclusions regarding the role of SBRT in patients with preserved lung function. One of the more compelling retrospective series reported on 87 medically operable patients with stage IA (n = 65) or stage IB (n = 22) NSCLC treated with SBRT where 5-year OS = 72 % for stage IA and 62 % for stage IB [19]. These results are very similar to surgical series for stage I NSCLC and have spawned the development of clinical trials to compare SBRT to surgery.