Stereotactic Radiosurgery for Small Intact Brain Metastasis: A Comparative Evaluation of 3 Different Single Fraction Prescription Doses

Abstract

While single-fraction stereotactic radiosurgery (SRS) in the treatment of small brain metastases (SBM, ≤ 2 cm) is well established, prescription dosing varies considerably across institutions and clinical trials. The choice of prescription dose is a delicate balance between local failure (LF) and radiation necrosis (RN) risks. In the modern era, historically-established dosing thresholds may no longer be applicable. To evaluate the impact of prescription dose on outcomes, we performed a comparative analysis of patients with SBM treated with definitive SRS using three different prescriptions, at a single tertiary institution. Consecutive patients with intact SBM treated with SRS from January 2017 and December 2021 were analyzed. Baseline patient characteristics and dosing parameters were abstracted from the medical record. To limit the integral brain dose when treating multiple brain metastases, the institutional practice was to reduce prescription dose as the total number of lesions increased (i.e., 24 Gy for ≤10 lesions, 22 Gy for 11-20, and 20 Gy for >20). A per lesion analysis, where each lesion was followed from the date of SRS to the last follow-up, was conducted with primary endpoints of LF and RN. Gray's test was used to compare the cumulative incidence of the LF and RN, with death as a competing risk. Factors affecting LF were analyzed using Cox hazard regression analysis. A total of 1318 SBM in 250 patients received SRS and met the inclusion criteria. The median age was 62 years (range: 18-90), median KPS was 90 (range: 50-100) and 66% were female. The most common primary tumors were lung (55.5%) and breast cancers (26.4%). With a median follow-up of 12 months, 136 (11%) LF in 44 patients and 70 (5.7%) RN events in 46 patients occurred. The actuarial 1-year cumulative rate of LF was lower in lesions treated with 24 Gy (6.4%, 95% CI: 4.7-8.6%) or 22 Gy (5.8%, 95% CI: 3.7-8.7%) compared to 20 Gy (15.4%, 95% CI: 10.9-20.5%) (p<0.01). 22 Gy and 24 Gy prescription doses were associated with a 44% and 52% reduction in risk in LF compared to 20 Gy (HR: 0.56; 95% CI: 0.36-0.9; p = 0.01 and HR: 0.48; 95% CI: 0.31-0.74; p<0.01, respectively). In a subset analysis of radiosensitive tumors, 1-year LF rate was still lower with 24 Gy (7.4%, 95% CI: 5.3-9.9%) and 22 Gy (6.1%, 95% CI: 3.7-9.4%) than 20 Gy (15.7%, 95% CI: 11.2-21%) (p = 0.01). The cumulative 1-year RN rate numerically declined with dose, but was not statistically significantly different, with 3.6% (95% CI: 2.3-5.3%) for 24 Gy, 2.6% (95% CI: 1.3-4.8%) for 22 Gy and 1.4% (95% CI: 0.4-3.7%) for 20 Gy. Patients treated with single fraction SRS to intact SBM were at increased risk of LF with prescription doses of 20 Gy compared to 22-24 Gy, without an increased risk of RN. Even in patients with radiosensitive histologies, higher LF rates were still observed following 20 Gy compared to 22-24 Gy.