1. Ferraro, Daniel MD, PhD
  2. Bradley, Jeffrey MD

Article Content

DANIEL FERRARO, MD, PHD (top), is Chief Resident in the Department of Radiation Oncology at Washington University School of Medicine and The Alvin J. Siteman Cancer Center in St. Louis, Missouri; and JEFFREY BRADLEY, MD, is Professor of Radiation Oncology, at Washington University School of Medicine and The Alvin J. Siteman Cancer Center in St. Louis, Missouri.

Figure. No caption a... - Click to enlarge in new windowFigure. No caption available.

ATLANTA-At the 2013 American Society for Radiation Oncology Annual Meeting, the topic of stereotactic body radiotherapy (SBRT) for patients with early-stage non-small-cell lung cancer (NSCLC) was addressed in multiple presentations. As this modality becomes more widely used in clinics across the country, we believe these reports are of interest to the larger oncology community. Summarized below are some of the reports we found of particular importance.


* Videtic and colleagues reported one-year results of RTOG 0915, a Phase II study comparing two different SBRT fractionation schedules in medically inoperable patients with stage I peripherally located (greater than 2 cm from the bronchial tree) NSCLC (Abstract 6). From September 2009 through March 2011, 94 patients with T1-2N0M0 disease were randomized to receive 34 Gy in a single fraction vs. 48 Gy total over four consecutive, once-daily 12 Gy fractions.



The primary endpoint of this study was grade 3 or higher toxicities at one year, with plans to select the most favorable regimen for a Phase III study comparing the three-fraction regimen defined by RTOG 0236. Grade 3+ toxicity at one year for the single-fraction regimen was 9.8 percent and 13.3 percent for the four-fraction regimen. Overall survival, a secondary endpoint, was not significantly different between the two regimens.


Although the follow-up is too short to change practice, these data suggest that a single fraction of SBRT for NSCLC may be as safe and effective as four fractions. Plans for a Phase III trial comparing one versus three fractions of SBRT for this population await longer follow up to confirm the safety and efficacy of single-fraction SBRT.


* Onishi and colleagues from Japan reported results of a large multi-institutional retrospective analysis of 2,226 patients who received SBRT for stage I NSCLC (Abstract 21). A wide range of doses (32-70 Gy), as well as number of fractions (3-12), was used in this cohort. The overall survival rate was 72 percent, and disease-specific survival was 85 percent at three years.



Grade 3+ NCI-CTC pulmonary complications occurred in 2.9 percent of patients. Subgroups who demonstrated better three-year overall survival rates included females (83% vs. 68%, p<0.01), patients who received doses with a biologically equivalent dose (BED) of 100 Gy or higher (75% vs. 63%, p<0.01), and medically operable patients (78% vs. 68%, p<0.01). Patients with pulmonary interstitial changes and/or emphysema had poorer overall survival.


* Hymas and colleagues from William Beaumont Cancer Institute reported results of a matched-pair analysis of patients with stage I NSCLC who received SBRT, lobectomy, or wedge resection (Abstract 23). A total of 286 patients treated between 2003 and 2010 were selected for matched-pair analysis. In head-to-head comparison, local recurrence at two and five years did not differ significantly between lobectomy and SBRT; however, local recurrence after wedge resection was significantly higher when compared with lobectomy.



With respect to overall survival at two and five years, lobectomy did significantly better than both wedge resection and SBRT. Regional and distant failure rates, as well as cause-specific survival did not differ significantly between the three groups.


These data suggest that SBRT is a reasonable alternative to wedge resection, while lobectomy remains the standard of care in operable patients. Additionally, the results suggest that SBRT may have similar rates of overall survival compared with lobectomy in a prospective randomized trial where patient selection bias can be minimized.


* Aneese and colleagues, also from William Beaumont, reviewed their SBRT experience in patients with tumors located centrally (less than 2 cm from the proximal bronchial tree) and compared toxicities in patients treated for peripheral tumors during the same time period (Abstract 80). Doses ranged from 48 to 60 Gy in four to five fractions (BED of 106-132 Gy).



In a matched-pair analysis, toxicity between peripheral and central tumors was similar, and severe toxicity was less than one percent. These results increase the anticipation for results of the RTOG 0813 Phase I/II multi-institutional trial. This trial used a five-fraction scheme to determine the maximum tolerable dose and local control with the use of SBRT in the treatment of proximally located early-stage NSCLC. Patients were treated to doses ranging from 10 Gy x 5 fractions to 12 Gy x 5 fractions.


The trial recently closed to accrual, in September 2013, with early results anticipated in one year.


* Finally, Shah and colleagues presented an analysis of the cost effectiveness of SBRT compared with surgical resection (either lobectomy or wedge resection) for stage I NSCLC (Abstract 98). Patients over 65 years old were assumed in the model, with both clearly operable and marginally operable scenarios investigated. The model takes into account disease, treatment, and toxicity data for SBRT, wedge resection, and lobectomy, and cost was calculated in 2012 dollars as well as quality-of-life adjusted years.



Lobectomy was found to be the most cost-effective option for clearly operable elderly patients, while SBRT was most cost effective for marginally operable patients over 65. With the National Lung Screening Trial showing a survival benefit for CT screening of high-risk patients, the number of early-stage NSCLC is likely to increase. Given this reality, further cost-effectiveness investigations comparing lung surgery with SBRT are likely.



All the abstracts referred to can be accessed via this link: