Intraoperative radiotherapy has the advantages of exposing the tumor bed directly to visual observation, optimizing the dose to complete radiotherapy in a single session, and avoiding radiation to normal tissue, thus improving the local outcome of radiotherapy. This technique is widely used in the treatment of a variety of tumors, however, the results are controversial. Initial reports by French and US research groups showed that intraoperative radiotherapy for breast cancer resulted in an increased rate of local recurrence compared to 50 Gy daily fractionated whole breast irradiation for more than 5 weeks plus 10C16 Gy external intensity-modulated tumor bed radiotherapy, which resulted in superior local tumor control rates and limited local recurrence to approximately 6% over a median follow-up period of 10 years. Although the risk of local recurrence is increased, intraoperative radiotherapy remains the preferred option for some patients because it eliminates the need to travel to radiotherapy centers for 25-33 divided whole breast radiotherapy sessions. In an issue of The Lancet, Jayant Vaidya and colleagues reported the results of the TARGIT-A trial, and in The Lancet Oncology, Dr. Umberto Veronesi and co-workers reported the results of the ELIOT trial. Each of these trials compared different forms of intraoperative radiotherapy with external whole-breast radiotherapy. The TARGIT-A trial set a non-inferiority cut-off value of 2.5% absolute difference in local recurrence rate between the two groups of breast-conserving patients. At the outset, intraoperative radiotherapy was performed in parallel with local mastectomy, but some medical centers also used intraoperative radiotherapy as a second step after obtaining a definitive tumor pathology report, with the aim of improving patient selection and allowing patients to be enrolled after local excision. Overall, the 5-year risk of local recurrence was 3.3% vs. 1.3% for breast-conserving patients in the intraoperative vs. whole-breast radiotherapy groups, respectively. These results are acceptable considering the prespecified non-inferiority boundary. Without pathological stratification, the risk of local recurrence was 2?1% vs 1?1% for patients in the intraoperative radiotherapy group versus the whole-breast radiotherapy group, respectively. This is a very important finding, which confirms the need for intraoperative radiotherapy with simultaneous local excision. In contrast, the ELIOT8 trial was not analyzed pathologically. The investigators concluded that intraoperative radiotherapy was considered equivalent to whole-breast radiotherapy if the local recurrence rate was less than 7.5%; the primary outcome was ipsilateral breast cancer recurrence (IBTR). The median follow-up period was 5.8 years, and the incidence of IBRT events was found to be 4.4% in the intraoperative radiotherapy group compared with 0.4% in the whole-breast radiotherapy group. Thus, the local recurrence rate in the intraoperative radiotherapy group fell within the predefined non-inferiority threshold, but was significantly worse than that in the all-breast radiotherapy group. Perhaps the most critical message about IBTR can be drawn from this: the authors were able to distinguish the true local recurrence rate from new ipsilateral breast cancer outside the index quadrant. True local recurrence, local recurrence outside the index quadrant, and axillary or local lymph node metastasis all increased significantly with intraoperative radiotherapy. In the ELIOT trial, there were 35 local recurrences in the intraoperative radiotherapy group, 14 (40%) of which occurred outside the index quadrant and 21 (60%) of which were true local recurrences. In the pathologically stratified TARGIT-A trial, local recurrence occurred in approximately 10 of 2234 patients. However, when examining the tumor characteristics of the TARGIT-A vs ELIOT trial, it was found that 12% versus 14% of tumor foci greater than 2 cm in diameter, 17% versus 26% of patients with positive lymph nodes, and 15% versus 20% of patients with grade 3 disease in each group, making it difficult to compare the results. In a multivariate analysis of the ELIOT trial, tumor size, presence of at least 4 positive lymph nodes, poorly differentiated tumors, and triple-negative subtypes were found to be associated with an increased incidence of IBRT. These data correlated with patient selection criteria in the phase 2 trial. The American Society for Radiation Oncology (ASTRO) Task Force recommendations confirm that the most relevant patient enrollment criteria for accelerated partial breast radiation are: age at least 60 years, tumor diameter less than 2 cm, stage T1N0 invasive ductal carcinoma, and estrogen receptor positivity. Patient selection may affect the final irradiation margin status. The lack of information on irradiation margins is the most criticized reason for intraoperative radiotherapy, and unfortunately, no exact description of irradiation margins was given in this trial. The greatest difficulty in comparing the results of the ELIOT and TARGIT-A trials was the difference between the two conventional whole breast radiation groups. In the ELIOT trial, whole-breast radiation included systematic 10 Gy intensity-modulated radiation, whereas the TARGIT-A trial did not specify the number of patients who received intensity-modulated radiation after whole-breast radiation. In the ELIOT trial, local recurrence occurred in only 4 of 654 patients (0?6%) in the whole-breast radiotherapy group, which was superior to the outcome of the EORTC trial with 16 Gy-modulated radiotherapy. After intraoperative radiotherapy, clear adverse histological features emerged that theoretically required further postoperative external radiation radiotherapy, which was performed in 22% and 5% of patients in the TARGIT-A and ELIOT trials, respectively. The effect of additional whole breast radiotherapy is difficult to interpret, as the ELIOT trial protocol specified additional irradiation only for patients with no less than four positive lymph nodes. The effect on overall survival varied depending on the number of lymph nodes involved, and nearly 75% of patients in the ELIOT trial received only adjuvant hormonal therapy, which is clearly inadequate if adjuvant chemotherapy is considered necessary for patients with hormone-receptor-positive tumors. In addition, patients in the ELIOT trial were defined as hormone receptor positive when they had more than 1% immune activated cells. Many other trials use a higher threshold, and the number of patients defined as hormone receptor negative is not negligible, and the administration of adjuvant chemotherapy would potentially lead to improved breast cancer mortality. Importantly, the median follow-up period of 2 years and 5 months for the TARGIT-A trial and 5.8 years for the ELIOT trial are too short to draw firm conclusions about the risk of death from breast cancer, especially for patients who do not meet the ASTRO recommendations but are enrolled. Expert commentary takes into account that without postoperative radiotherapy, the incidence of IBRT increases over time and that patients have a sufficiently long life expectancy after diagnosis that recurrence and disease metastasis may occur, so we do not support neglecting radiotherapy after local excision. Intraoperative radiotherapy has few side effects and therefore may be an attractive option for some low-risk patients. One way to improve the acceptability of external radiation radiotherapy is to reduce the number of fractions or radiation dose, if possible to reduce the total treatment time. Studies with a median follow-up of at least 10 years have confirmed the safety and effectiveness of large-split radiotherapy for breast cancer. This clinical pathway is increasingly used in the clinic, but only for patients who are unable to undergo extended radiotherapy or intraoperative radiotherapy. It is more commonly used in patients aged 55 to 65 years with tumors less than 2 cm in diameter and negative lymph nodes. Large fractionated radiotherapy can provide greater convenience to patients and save them a lot of waiting time in line.