Abstract Introduction. Breast-conserving surgery (BCS) followed by postoperative radiotherapy (RT) still represents the standard of care for early breast cancer (BC) patients. Hypofractionated schedules in maximum 15 fractions are currently accepted as the gold standard for external beam whole and partial breast irradiation (PBI). PBI for selected early BC patients allowed a shorter overall treatment duration and an improved patient compliance as compared to old-fashioned RT schedules. Preoperative PBI, due to the advantage of treating a well-defined volume, has been gaining attention in this multidisciplinary scenario. It avoids local treatment delay and might allow tumour downstaging with increased rates of BCS and improved cosmetic outcomes. Since local recurrence might be driven by biological mechanisms of radioresistance rather than geographical miss, higher dose per fraction may overcome repair mechanisms allowing tumoral cells to escape from conventional RT damage. We report the results of the phase II ROCK trial (NCT03520894), enrolling early BC patients treated with preoperative robotic radiosurgery (prRS), in terms of acute and early late toxicity, disease control, and cosmesis. Material and methods. The study recruited between August 2018 and September 2021 at the Radiation Oncology Unit of the University of Florence (Florence, Italy). Eligible patients were women aged 50+ years, with histologically proven invasive early BC, HR+/HER2- disease, without lymph vascular invasion, tumour size up to 25 mm suitable for BCS. Exclusion criteria were clinical node positive disease, multiple foci tumours, and patients with breast lesion limiting within 5 mm from the skin surface. The study aimed to prospectively assess the safety and feasibility of a single Cyberknife® (Accuray Incorporated, Sunnyvale, CA, USA) 21Gy-fraction prRS in preoperative setting, and to identify predictive factors for outcomes based on biologic and clinical features. The primary endpoint was the acute skin toxicity (from the end of prRS to surgery) according to the RTOG and the EORTC scales. Secondary endpoints were the rate of early late skin and non-skin toxicity as measured 90 days from the end of prRS, the rate of pathological complete response (pCR) according to Chevalier score. Cosmetic outcomes were prospectively scored every 6-month using the BCCT.core software. Results. From August 2018 to September 2021, a total of 70 patients were recruited and enrolled. Of those, 41 were excluded due to tumour biology exclusion criteria and 7 due to multiple foci breast disease evidenced at basal MRI. Therefore, 22 patients were successfully treated with pRS. Median age at diagnosis was 68 years (range 50-86), median tumour size was 14 mm (range 7.5-25). Required target dosimetric parameters were met in all patients, as well as normal tissue constraints. Patients received surgery after a median time of 29 days from biopsy, without any delay or postoperative complication. Overall, three G1 adverse events (13.6%) were recorded within 7 days from prRS (1 erythema, 2 breast pain). Three events (13.6%) were recorded between 7 and 30 days from prRS, one G2 breast oedema and two G1 breast pain. No acute toxicity greater than G2 was recorded. Five patients experienced early late G1 toxicity (1 breast pain, 4 breast induration). One patient reported G2 breast induration. No early late toxicity greater than G2 was observed. At a median follow up of 18 months (range 6-29.8), cosmetic results were scored excellent/good and fair in 14 and 5 patients, respectively, while 3 patients experienced a poor cosmetic outcome. Overall, pCR after surgery was reported in 2 patients (9%). Two patients received postoperative whole breast irradiation, according to histopathological results. Conclusions. ROCK trial showed that a single 21 Gy dose prRS represents a feasible technique for selected patients affected by early BC, showing a good safety profile and a promising effectiveness. Citation Format: Icro Meattini, Giulio Francolini, Vanessa Di Cataldo, Luca Visani, Carlotta Becherini, Erika Scoccimarro, Monica Mangoni, Viola Salvestrini, Laura Masi, Chiara Bellini, Raffaela Doro, Federica Di Naro, Marco Bernini, Jacopo Nori, Lorenzo Orzalesi, Simonetta Bianchi, Lorenzo Livi. Preoperative robotic radiosurgery for early breast cancer: results of the phase II ROCK trial (NCT03520894) [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr PD3-03.