Background: A good cosmetic outcome has been defined as an important endpoint in breast-conserving therapy (BCT). Various evaluation methods have been studied, but the optimal method has yet to be identified. The present supplementary analysis of JCOG0906 focused on comparing evaluation methods for breast cosmetic outcomes following hypofractionated whole breast irradiation (HFWBI) to examine whether a computer-software (the Breast Cancer Conservative Treatment cosmetic results [BCCT. core])-based program evaluation (CE) can be used for Asian women in clinical trials of BCT.
Methods: Of 306 women, 292 underwent institutional evaluation (IE) for breast cosmetic outcomes before (pre) and 3 years after (post) HFWBI using a 4-point scale (excellent/good/fair/poor), and they were evaluated by CE and a central panel evaluation (PE) on the same scale using 292 pairs of pre/post-HFWBI photographs. PE was performed twice by consensus of the same two experts with a 3-year interval. CE was assessed individually by two radiation oncologists, an expert and a non-expert. Intra-observer variability and inter-observer variability were calculated using the kappa (k) and weighted kappa (wk) statistics.
Results: The agreement between the first and second PE using pre/post-HFWBI photographs was moderate (k = 0.60, wk = 0.64. k = 0.53, wk = 0.60). The agreement between the expert and non-expert on CE was substantial (k = 0.72, wk = 0.76. k = 0.72, wk = 0.77). The inter-observer variability of CE was smaller than the intra-observer variability of PE.
Conclusion: CE with BCCT. core was considered a reproducible and an appropriate evaluation method for Asian women in clinical trials of BCT, when breast cosmetic changes were compared between pre/post therapy.
Keywords: BCCT. core software; Breast cosmetic evaluation; Breast-conserving therapy; Hypofractionated whole breast irradiation.
© 2022. The Author(s), under exclusive licence to The Japanese Breast Cancer Society.