2796 - Dose-LET Interactions Predict Capsular Contracture after Proton Postmastectomy Radiation Therapy
Presenter(s)
Z. Ma1, J. Chen2, M. Cao2, R. W. Gao1, Y. Yang3, Y. Ding2, N. Remmes1, J. Ma1, K. S. Corbin1, D. Shumway1, W. Liu2, and R. W. Mutter1; 1Department of Radiation Oncology, Mayo Clinic, Rochester, MN, 2Department of Radiation Oncology, Mayo Clinic, Phoenix, AZ, 3University of Miami, Miami, FL, United States
Purpose/Objective(s): Proton postmastectomy radiation therapy (PMRT) has been increasingly adopted for its dosimetric advantages. Recent single institutional studies have suggested that breast implant reconstruction may be vulnerable to capsular contracture following proton therapy, which has been hypothesized to be related to the higher proton linear energy transfer (LET). This study aimed to investigate the combined effects of dose and dose-averaged linear-energy-transfer on capsular contracture after proton PMRT, and to derive clinically interpretable dose–LET volume constraints.
Materials/Methods: Patients with breast cancer who underwent implant reconstruction and proton PMRT (50 Gy in 25 fractions) at our institution between 2015 and 2021 were included. Patients with capsular contracture were matched 1:2 with controls using nearest-neighbor matching. We defined a novel organ-at-risk (OAR), peri-implant tissue, as a 5-mm shell surrounding the implant for proton therapy. The dose-LET volume histogram (DLVH) was calculated. DLVH is a multidimensional extension of the traditional dose volume histogram, which characterizes the normalized volume of a structure as a simultaneous function of both physical dose and dose-averaged LET. Generalized linear mixed-effects regression was employed to identify DLVH indices significantly associated with capsular contracture. Spearman correlation analysis was used to eliminate redundant DLVH indices. Dose–LET volume constraints were derived from receiver operating characteristic analysis and validated using a support vector machine-based normal tissue complication probability (NTCP) model with leave-one-out cross-validation.
Results: 145 patients treated with conventionally fractionated proton therapy were included, of these, 8 patients with documented capsular contracture were identified and matched with 16 control patients without contracture. Baseline characteristics were well-balanced between the two cohorts. Three independent DLVH indices were found significantly associated with capsular contracture (p<0.01): V(55.8 Gy, 2.2 keV/µm), V(50.3 Gy, 5.4 keV/µm), and V(32.8 Gy, 0.9 keV/µm). The corresponding DLVCs were V(55.8 Gy, 2.2 keV/µm) < 0.0033%, V(50.3 Gy, 5.4 keV/µm) < 0.0017%, and V(32.8 Gy, 0.9 keV/µm) > 96.98%. The NTCP model achieved an AUC of 0.867, with an accuracy of 91.7%, sensitivity of 87.5%, and specificity of 93.8%.
Conclusion:
Capsular contracture following proton PMRT is significantly associated with the joint interplay between physical dose and LET in peri-implant tissue. The derived dose–LET volume constraints can be integrated into proton treatment planning using a novel OAR definition to reduce capsular contracture risk. These findings warrant validation in larger prospective cohorts before clinical implementation.