2731 - Postoperative Breast Radiation Therapy Using Deformable Image Registration of Initial PET/CT in the Same Posture before Primary Systemic Therapy: Does PET-Based Precise Planning Translate into a Clini

K. Shiraishi¹, A. Oyama², S. Kumagai³, R. Uehara³, A. Nomoto¹, J. Kotoku¹, and H. Jinno¹; ¹Teikyo University, Tokyo, Japan, ²Osaka University, Osaka, Japan, ³Teikyo University Hospital, Tokyo, Japan

Purpose/Objective(s): Radiation treatment planning (RTP) based on both the clinical stage at diagnosis and the pathological stage after surgery is highly recommended in postoperative radiation therapy (RT) for patients with breast cancer who undergo primary systemic therapy (PST). We previously reported initial PET/CT deformable image registration (DIR) technique to solve inconsistency due to different acquisition conditions for precise RTP, by which the coverage of affected axillary lymph nodes at diagnosis was significantly improved. Here we present relevant clinical outcomes in patients who underwent PET/CT under the same conditions as those at the RTP (flat panel on the back and both upper arms raised, group A) and those who did not (group B). Materials/

Methods: Consecutive patients in Group A underwent PET/CT after May 2019 (N=395) and patients in Group B had got PET/CT before then since 2016 (N=78), irrespective of following RTP according to our strategy in PST settings. Regions of interests were generally set according to the RTOG contouring guideline, with appropriate individual modifications. All primary tumors and lymph node metastases considered to be positive at initial PET/CT before PST were included in the target. Cases in which PET/CT was taken with both arms raised, but appropriate target coverage after DIR was not conducted in review, were designated as group B. Prescribed doses were 42.56-50 Gy in 16-25 fractions. Boost irradiation was added to indicated cases. The primary outcome was regional lymph node control. Secondary outcomes were distant disease control, relapse-free survival, and disease-free survival. Duration of survival was defined as the time interval between the date of initiation of RT and the date of events or the censor date of the last follow-up. We estimated propensity scores with a logistic regression model for the likelihood of undergoing PET/CT in Group A as a function of all variables of patient characteristics including age, clinical stage at diagnosis, pathologic stage, tumor location (inner/outer), boost irradiation (yes/no), nuclear grade, biological subtype, and Ki-67 values before and after PST.
Results: Comparison by Cox proportional hazards model after inverse probability of treatment weighting (IPTW) with propensity scores to adjust the balance between the two groups showed significant differences not only in regional (adjusted HR [AHR], 0.095, [95% CI, 0.017-0.543]) and distant disease control (AHR, 0.190, [95% CI, 0.069-0.524]), but also in disease-free survival (AHR, 0.350, [95% CI, 0.175-0.698]). Overlap weighting (OW)-adjusted Cox regression model also supported the benefit of our sophisticated approach for disease control. We did not find any meaningful difference in toxicity in terms of lymphedema, though more >grade 2 pneumonitis were seen in Group A.
Conclusion: RTP using initial PET/CT under the same conditions as RT can yield target coverage improvement, and in addition, better short-term prognosis by IPTW- and OW-adjusted analyses.