Room: Track 3
Purpose: Emerging clinical evidence supports the variability of relative biological effectiveness (RBE) in proton radiotherapy. This poses the need to account for RBE variability in proton planning. However, no harmonized concept exists on how to calculate the RBE-driving linear energy transfer (LET) in clinical practice. Therefore, a multi-centric study was set up with the objective to standardize clinical LET calculations in Europe.
Methods: Eight European institutions generated non-robust SOBP plans using common strict dose objectives. Multiple treatment field arrangements (single-field SOBP, perpendicular fields, opposing fields) were employed to cover a target cube in a water phantom. Each institution used its preferred treatment planning software and provided dose and corresponding LET distributions for a joint analysis.
Subsequently, RBE-weighted dose (DRBE) distributions were calculated for the single-field SOBP of one institution assuming the Wedenberg RBE model using Monte Carlo calculated unrestricted dose- and track-averaged LET (LETd/LETt) distributions considering (1) only primary protons, (2) all protons, (3) all particles with Z=2.
Results: Institutional SOBP ranges and target average doses agreed within 2%. In contrast, near-minimum, average and near-maximum LETd differed up to 30%, 19% and 5% in the target, respectively. These discrepancies could partially be explained by different algorithms (Monte Carlo/analytical) and by different ions included in the LETd calculations.
LETd calculations were more sensitive to the considered secondary particle spectrum than LETt. Deriving DRBE using LETd yielded 0-11%, 4-12% and 12-45% higher DRBE in the entrance, target and distal edge regions, respectively, compared to LETt. The biological range extension using LETd (and LETt) was approximately 3 mm (and 1 mm).
Conclusions: Despite comparable dose distributions, substantial LET differences occurred among the participating institutions. These differences hamper the consistent analyses of clinical follow-up data as they translate to substantial discrepancies in predicted DRBE. Therefore, standardization of clinical LET calculations is of utmost importance.