3D conformal external beam radiotherapy (3D-CRT) for accelerated partial breast irradiation (APBI): What is the correct prescription dose

Abstract

APBI has traditionally been delivered with interstitial brachytherapy, with 3D-CRT emerging as a promising technique. The most widely used fractionation scheme for APBI using HDR brachytherapy consists of 34 Gy delivered in 10 fractions over five days, and 3D-CRT has been introduced utilizing 38.5 Gy delivered in 10 fractions over five days. The dose prescription used for breast brachytherapy is the current standard in the delivery of APBI. No literature currently exists to confirm the equivalence of the dose prescription for 3D-CRT. This study is a case-based evaluation of the biological equivalence of the dose prescriptions for brachytherapy and 3D-CRT APBI, based on the concept of equivalent uniform biologically effective dose (EUBED). Both 3D-CRT and multi-catheter brachytherapy plans were generated for each of five patients. A CTV and PTV were created for each patient per the current APBI protocol guidelines, with PTVs ranging from 131 to 282 cm3. Dose matrices for the brachytherapy and 3D-CRT plans were computed using the Varian Brachyvision and Pinnacle planning systems, respectively. These dose matrices were then exported and analyzed with in-house software. The EUBED is used in this study to compare a highly non-uniform dose distribution of a given total dose and fraction size (brachytherapy) with a more homogeneous one (3D-CRT). The EUBED is a combination of equivalent uniform dose (EUD) and biologically effective dose (BED), allowing a non-uniform dose distribution to be reduced to an equivalent uniform dose while also accounting for fraction size. For each patient, the biologically effective dose (BEDi) was calculated for each voxel in the brachytherapy dose matrix using the standard linear quadratic (LQ) model, with α=0.3 Gy−1 and β=10 Gy. These values were then used to generate the equivalent uniform biologically effective dose for the entire brachytherapy dose matrix (EUBEDbrachy). Since EUBED is not a value easily related to in terms of clinical experience, it can be converted into a homogeneous EUD delivered at a given dose per fraction (eud). Using n = 10 fractions, one can calculate the EUD and eud that will produce the same EUBED. The equivalent uniform dose for the 3D-CRT plans (EUD3D-CRT) was then calculated from EUBEDbrachy, with the constraint that the treatment should be delivered with the same fractionation schedule. The generalized EUD (EUDa) was calculated for breast tissue with local control as the end-point (a = −7.2) for both the HDR and 3D-CRT distributions. Despite large differences in the geometry of the implants for the five patients studied, the values of generalized EUDa for the PTV were remarkably similar (39.4 ± 0.2 Gy). For the larger volume enclosed by a lower isodose line (e.g. 80%), the calculated EUDa was closer to the prescription dose (34.3 Gy). The calculated values of EUBEDbrachy were also quite consistent (51.5 ± 0.4 Gy). Using the values of EUD3D-CRT derived from EUBEDbrachy, we obtained a mean fraction size of 3.74 ± 0.03 Gy, and an equivalent prescription dose of 37.4 Gy. The generalized EUDa computed for dose distributions which were actually delivered with 3D-CRT was slightly higher, averaging 39.4 Gy, but does not account for the effect of different fraction sizes. The published experience suggests that the presently accepted dose prescription for breast brachytherapy yields excellent local control. Utilizing established radiobiological parameters, this study suggests that the ideal fraction size needed to deliver a biologically equivalent dose using 3D-CRT is at least 3.74 Gy. Therefore, the presently employed fraction size of 3.85 Gy for 3D-CRT appears to be appropriate to achieve equivalent local control