1079 - Effect of Radiation Dose on Lymphedema Development following Immediate Lymphatic Reconstruction and Axillary Lymph Node Dissection
Presenter(s)
D. Spiegel1, J. Levey1, M. Keko2, A. M. Modest3, J. N. N. Ntambi1, R. Friedman4, D. Singhal5, and A. Recht1; 1Department of Radiation Oncology, Beth Israel Deaconess Medical Center, Boston, MA, 2Department of Orthopedics, Beth Israel Deaconess Medical Center, Boston, MA, 3Department of Obstetrics and Gynecology, Beth Israel Deaconess Medical Center, Boston, MA, 4Department of Plastic Surgery, University of California Los Angeles, Los Angeles, CA, 5Division of Plastic and Reconstructive Surgery, Department of Surgery, Beth Israel Deaconess Medical Center, Boston, MA
Materials/Methods:
This prospective study included 23 patients with invasive breast cancer who underwent ALND/ILR followed by breast/chest wall RT and RNI from 10/2020-6/2023. A twirl clip placed at the anastomosis allowed for contouring of the ILR site during RT planning. The median breast/chest wall RT dose was 4500 cGy/18 fractions; the median RNI dose was 4000 cGy/16 fractions. All patients received RT to axillary Level 3 and the supraclavicular fossa; Levels 1-2 were targeted in 14 patients. LE was defined by a 10-point increase in LDEX in the dominant affected extremity, 7-point in the non-dominant extremity, or 10% increase in arm volume compared to pre-ILR baseline plus patient-reported symptoms occurring > 6 months after RT. Dosimetric parameters included mean and maximum doses, V35, V40, and Dmin<36.8 Gy at the ILR site, ILR + 5 mm, and ILR + 2 cm expansion volumes. Data are presented as median and interquartile range (IQR) or n (%). Categorical data was compared using Chi-square or Fisher’s exact tests. Continuous data was compared using Mann-Whitney-U test. Poisson regression with robust error variance was used to calculate the risk ratio and 95% confidence interval (CI). ROC curves and thresholds for the dosimetric parameters were calculated based on the Youden index and the area under the curve (AUC). Results: Median follow-up was 23.4 months (IQR, 20.1-29.2). The median age was 56 years (45.5-64.0). Median BMI at surgery was 26.4 kg/m2 (24.4-31.4). 87% had > 10 lymph nodes removed. 14 patients met criteria for lymphedema at a minimum of 1 time during the study period, but only 3 (13%) met criteria at the end of the study period. Patients who developed LE had higher mean dose (4135 cGy vs 1410 cGy, p=0.006), V35 (89% vs 20%, p=0.005), and V40 (89% vs. 17%, p=0.012) at the ILR + 2 cm volume compared to those that did not. These parameters for the ILR + 2 cm volume remained significant predictors of LE when controlling for BMI and lymph nodes removed: mean dose, relative risk [RR] 1.465 (95% CI 1.054-2.035); V35, RR 1.185 (95% CI 1.026-1.369); and V40, RR 1.149 (95% CI 1.015-1.302). Threshold doses above which the risk of LE increased significantly were found for the ILR + 2 cm expansion: mean dose, 3074 cGy (AUC 0.86), with rates of LE above and below the threshold of 92% vs 30%, p=0.006; V35, 56% (AUC 0.87), 92% vs 22%, p=0.001; and V40, 50% (AUC 0.83), 92% vs 30%, p=0.006. Conclusion: Increased RT dose to the ILR anastomosis site and the volume surrounding it increases the risk of LE. Future studies will assess whether limiting dose to this region using these newly defined threshold doses can lower BCRL rates while maintaining disease control.