Dosimetric Parameters are Associated with Local Control after SBRT for Colorectal Cancer Lung Metastases

Abstract

<h3>Purpose/Objective(s)</h3> Recent data suggest that higher stereotactic body radiotherapy (SBRT) prescription doses, and hence biologically effective doses (BED), are associated with improved local control (LC) for patients who receive SBRT for colorectal cancer (CRC) oligometastasis. However, the immense heterogeneity of SBRT dose distributions, even at isoeffective prescription doses, has made it challenging to extrapolate reported prescription dose thresholds into contemporary plan optimization. Our objective was to characterize dosimetric predictors of LC after SBRT for CRC lung metastases. We hypothesized that the internal target volume (ITV) and planning target volume (PTV) doses covering 90% of the volume (D90) and minimum doses (Dmin) would be associated with local control. <h3>Materials/Methods</h3> This was a retrospective cohort study including patients who received SBRT to CRC lung metastases at a single cancer center between 2013-2019. Dosimetric parameters including the D90 and Dmin of both the ITV and PTV were collected and converted to BED using the linear quadratic model assuming an alpha/beta of 10 Gy. The Kaplan-Meier method was used to estimate LC, and univariate analyses were performed with the Cox proportional hazards regression model. <h3>Results</h3> 53 CRC lung metastases amongst 28 patients were included for analysis. Median dose and number of fractions were 50 Gy (IQR 50-54 Gy) and 5 (IQR 3-5) fractions, respectively, with a median BED of 100 Gy (IQR 100-151.2 Gy). Median follow-up after SBRT was 27 months (IQR 17 – 38 months). By last follow-up, 18 of 53 (34%) metastases had progressed, of which only 4 (22%) were isolated sites of first-progression. The 2-year LC was 66.0% (95% CI 54.3% – 81.6%). On univariate analysis, lung metastasis size (HR 1.601 per cm, 95% CI 1.052 – 2.436) and BEDs including the ITV D90 (HR 0.83 per 10 Gy, 95%CI 0.75–0.93, p = 0.002) and Dmin (HR 0.821 per 10 Gy, 95% CI 0.73–0.93, p = 0.002), and PTV D90 (HR 0.79, 95%CI 0.68–0.93, p = 0.004) and Dmin (HR 0.80, 95% CI 0.66–0.97, p = 0.025) were associated with LC. When BED dose thresholds were chosen based on bottom quartile, improved 2-year LC was associated with ITV D90 BED ≥118 Gy (84.6% vs. 21.4% p < 0.0001) and Dmin BED ≥ 114 Gy (81.5% vs. 30.5% p < 0.0001). This remained true among the 29 lesions treated with the most common prescription dose of 50 Gy in 5 fractions (90.0% vs. 22.2%, p < 0.0001 and 83.5% vs. 37.5%, p = 0.011, respectively). <h3>Conclusion</h3> BEDs of over 118 Gy to at least 90% of the ITV volume (D90) and a minimum dose of 114 Gy were associated with improved LC after SBRT for CRC lung metastases. These data may be informative in-patient selection and plan optimization, particularly in challenging circumstances where the target is adjacent to critical organs at risk. With the favorable LC achieved with dose-escalated SBRT and the infrequency of isolated in-field first-progression events, SBRT remains an excellent local treatment option for CRC lung metastasis.