A nomogram for predicting in-breast tumor recurrence risk in breast cancer patients treated with partial breast irradiation using intraoperative electron radiation therapy.

[PURPOSE] The study aims to develop and validate a predictive tool for assessing the risk of in-breast tumor recurrence (IBTR) in breast cancer patients considered candidates for intraoperative radiotherapy using electrons (IOERT).

[METHODS] This study included 3397 breast cancer patients treated with IOERT at a single institution between 2000 and 2016. The primary endpoint was IBTR, with or without nodal or distant metastasis. Fine and Gray regression models were used to identify predictors of IBTR. A nomogram predicting the 5- and 10-year probability of IBTR was developed based on the multivariable model and was validated both internally and externally using data from the IOERT arm of the ELIOT phase III trial (585 patients).

[RESULTS] With a median follow-up of 6.1 years (interquartile range 4.3-8.0), 265 IBTRs (7.8%) were observed, resulting in an IBTR cumulative incidence of 4.4% (95% CI 3.7-5.2) at 5 years and 13.5% (95% CI 11.7-15.5) at 10 years. Multivariable analysis revealed that age under 60, certain histologic subtypes, positive axillary nodes, and intermediate/high tumor grade were key risk factors for IBTR. The overall Harrell's concordance statistic was 0.69 (95% CI 0.66-0.73) in the internal and 0.64 (95% CI 0.57-0.71) in the external validation.

[CONCLUSION] The nomogram has demonstrated moderate discriminative ability in predicting IBTR in the internal validation set and may be a useful tool to support treatment decision-making in breast cancer patients eligible for IOERT.