2546 - Single-Fraction vs. Multi-fraction Stereotactic Ablative Body Radiotherapy [SBRT] for Colorectal Pulmonary Oligometastases

J. Kocsis¹, C. A. Reddy², C. S. Shah³, J. A. Miller⁴, S. R. Amarnath², E. H. Balagamwala², G. M. Videtic⁵, and K. L. Stephans⁵; ¹Cleveland Clinic, Cleveland, OH, United States, ²Department of Radiation Oncology, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, ³Department of Radiation Oncology, Cleveland Clinic Foundation, Cleveland, OH, ⁴Cleveland Clinic, Cleveland, OH, ⁵Department of Radiation Oncology, Cleveland Clinic, Cleveland, OH

Purpose/Objective(s): SAFRON II was a randomized phase II trial testing single vs multi-fraction (fx) lung SBRT in patients (pts) with pulmonary oligometastases (OM). Its primary endpoint was toxicity, and secondary endpoints included local control (LC) and overall survival (OS). It showed no significant difference between the study arms in terms of freedom from local failure (LF) and OS. However, a post hoc exploratory analysis suggested LF in the single-fx arm was higher for patients with colorectal (CRC) primaries. We conducted a retrospective review (RR) of our institutional experience treating CRC pulmonary OM with single and multi-fx SBRT with respect to LC and OS. Materials/

Methods: We queried an IRB-approved lung SBRT registry for pts with peripheral lung CRC OM lesions between 2005-2023. Pts were assessed for LC, OS, and toxicity. Rates of LF were calculated using cumulative incidence analysis with death treated as a competing event, and rates of OS were calculated using actuarial analysis. Competing risk regression was used to identify factors associated with LF.
Results: 78 pts met eligibility criteria and 133 lesions were treated. 35 pts (44.9%) were alive at time of analysis. Median clinical follow-up (fu) time was 27.9 months (mos) (15.1 mos for single fx and 40.4 mos for multi-fx). The median age was 62.4 years (range 42.5-88.5), 53.8% were female, and median KPS 90 (range 60-100). 32 (41.0%) pts had multiple lesions treated and median number of synchronous lesions treated was 3 (range 2-5). Median size was 1.1 cm (range 0.3-4.0). KRAS status was known in 61 pts (78.2%) and 41% of all pts were KRAS mutated. 40 lesions (30.1%) were treated with single fx (30 Gy/1 fx [n=0.8%], 34 Gy/1 fx [n=29.3 %]) and 93 lesions (69.1%) were treated with multi-fx (50 Gy in 5 fx [24.8%], 60 Gy in 5 fx [20.3%], and 60 Gy in 3 fx [15.0%]). Cumulative incidence of LF at 12 months was 5.9% (95% CI 2.6-11.1) for all lesions, and 6.6% (95% CI 2.7-13.0) in the multi-fx group vs 3.4% (95% CI 0.2-14.9) in the single fx group (p=0.70). On univariate analysis, single fx was not predictive of LF (HR 1.40; 95% CI, 0.312-6.29; p=0.13), and tumor size was predictive of LF (HR 1.83, 95% CI, 1.08-3.09, p=0.03). When stratified by tumor size (=1 vs >1 cm), fractionation was not significant for LF for tumors >1 cm (HR 0.97; 95% CI, 0.12-7.9) or for tumors =1 cm (HR 1.17; 95% CI, 0.12-12.0). KRAS mutation did not predict for LF (HR 1.28, 95% CI 0.37-3.47, p=0.70). Median OS for the multi-fx arm was 68.9 mos and median survival for the single fx arm was not reached. There were 15 total adverse events reported. Grade 1-2 chest wall toxicity was reported in 9 (6.8%) lesions- 6 (15%) in single fx vs 3 (3.2%) in multi-fx. Grade 2 pneumonitis occurred in 5 lesions (3.8%)- 2 (5.0%) in single fx and 3 (3.2%) in multi-fx.
Conclusion: This RR supports single fx SBRT to 34 Gy in 1 fx in the treatment of CRC pulmonary OM given its efficacy and safety. Larger lesions may be more likely to fail locally independent of fractionation. KRAS status was not associated with LF in this analysis.