Cold-spot driven local failure after stereotactic body radiation therapy for colorectal liver metastases

This single-institution retrospective study investigated whether actual dosimetric plan quality metrics better predict freedom from local progression (FFLP) than nominal prescription dose after stereotactic body radiation therapy (SBRT) for colorectal liver metastases (CLM).

A total of 116 patients with 128 CLM treated with SBRT (≥70 Gy biologically effective dose [BED₁₀]) were retrospectively analyzed. Clinical variables and dosimetric factors, including maximum, minimum, and mean doses (Dmax, Dmin, Dmean), as well as dose to x% of the planning target volume (PTV) (Dx%), were evaluated. Cox regression analyses and the Akaike information criterion were used to identify prognostic factors for FFLP.

With a median follow-up of 22.3 months, 1-year and 2-year FFLP rates were 74.5% and 58.8%, respectively. On univariable analyses, extent of metastasis, number of systemic therapy lines, pre- and post-SBRT carcinoembryonic antigen, PTV, prescription dose, and other plan parameters (Dmax, Dmin, Dmean, D2%, D95%, D98%, D99%) were significant. In the final multivariable model, PTV Dmin (HR 0.89 per 10 Gy BED₁₀, 95% CI 0.83-0.96, p = 0.002) and polymetastatic disease (HR 3.31, 95% CI 1.79-6.12, p < 0.001) independently predicted FFLP. Using an optimal cut-point of 100 Gy BED₁₀, low-PTV Dmin was associated with inferior FFLP compared with high-PTV Dmin (1-year: 64.2% vs. 87.1%, p < 0.001).

PTV Dmin emerged as the most robust dosimetric predictor of FFLP. Incorporating PTV Dmin into SBRT planning may improve FFLP beyond nominal prescription dose.