Introduction: Sentinel lymph node biopsy (SLNB) is integral to breast cancer surgery, providing accurate pathological staging that guides treatment and prognosis. Magtrace® is a novel superparamagnetic tracer for sentinel lymph node (SLN) localisation that offers flexible injection timing and avoids reliance on nuclear medicine. Magtrace® provides comparable SLNB performance to Technetium99 (Tc99) while reducing preoperative workflow requirements. We aim to prospectively compare Magtrace® with the standard Tc99 for localisation of SLN, including clinical outcomes and procedural complications. Methodology: Prospective data collection for patients undergoing Magtrace® injection for SLNB for breast cancer surgery was carried out from March 2023 to December 2025. Consented patients undergoing surgery who met selection criteria underwent injection of Magtrace® prior to their surgery. Results: 61 participants underwent 62 Magtrace® injections, (time range: 0-29 days preoperatively). Surgeries included 57 mastectomies and 5 wide local excisions. SLNB were undertaken in 58 cases, including 52 immediate SLNBs and 6 delayed SLNBs (dSLNBs); 4 participants did not undergo SLNB. Of the 58 SLN procedures, 50 procedures utilised both Magtrace® and Tc99, while the remaining 8 used Magtrace® alone. Using dual-tracer, Magtrace® identified more SLNs than Tc99 in 7 cases. SLN yield ranged from 1 to 5 nodes for both techniques, with mean yields of 1.9 for Magtrace® and 1.72 for Tc99. Grey skin discolouration occurred in 26 participants; no other adverse effects were observed. Conclusion: Magtrace® is an effective alternative to Tc99 for SLN localisation, providing similar node yield and flexible injection timing. Its use may improve access to SLNB in centres without nuclear medicine support and facilitate safe delayed SLNB when invasive disease is identified on final histology.

Purpose: Breast cancer–related lymphoedema (BCRL) affects up to 30% after axillary lymph node dissection (ALND) and is difficult to treat, highlighting the need for preventive strategies. Lymphaticovenous anastomosis (LVA) offers a potential solution when performed at index operation. Indocyanine green (ICG) fluorescence axillary reverse mapping (ARM) has been shown to be superior to blue dye ARM, with 95% accurate lymphatic mapping based on the authors’ prior work. This study reports early technical and clinical outcomes from the first ten cases. Methodology: Consecutive patients with node‑positive breast cancer undergoing ALND received ICG fluorescence‑guided ARM followed by immediate LVA at a single institution (April–August 2025). ARM lymphatics were identified with ICG and microsurgically anastomosed to a venous tributary. Anastomotic patency was assessed intraoperatively using ICG transit. Operative details, complications, and lymphoedema outcomes (bioimpedance spectroscopy and Norman Questionnaire) were assessed at 6 months. Trial registration: ACTRN12625000414415. Results: Ten patients were included. Nine underwent a single technically successful LVA; one patient lacked a suitable recipient vein. Anastomoses used an intussusception technique, comprising eight end‑to‑end anastomoses (one requiring a jump graft) and one arborised anastomosis. Patency was confirmed intraoperatively in all successful cases using ICG. Median duration of LVA component was 50 mins (IQR 40–95). One postoperative seroma occurred; no infections or anastomosis‑related complications were observed. At 6‑month followup, no patients had developed lymphoedema. Conclusion: This pilot demonstrates feasibility and safety of ICG‑guided LVA during ALND, with high technical success, reliable intraoperative patency confirmation, and no early lymphoedema. These findings support longer‑term followup within the full prospective trial to assess durability and lymphoedema risk reduction.

Purpose: Magseed localisation is increasingly popular in breast surgery due to its ability to be inserted any time pre-operatively without signal decay, lack of ionising radiation, ease of operator use, and low rates of migration. We reviewed the outcomes and complications using this novel technique for excision of non-palpable breast lesions and targeted axillary lymph nodes. Methods: Prospective data was collected from June 2022 to December 2024 in 500 patients who had Magseed localisation of breast and axillary lesions in a tertiary Australian metropolitan hospital. Results: A total of 555 Magseeds were used in women aged 22 to 87 years old. The majority had a single breast lesion localised with a single Magseed (n=473), with multiple Magseeds for multiple lesions within the same breast in 14 cases. Bracketing was used in 20 cases. Targeted axillary dissection was performed for 13 patients. There were 363 wide local excisions (n=326 level 1 oncoplasty, n=37 level 2 oncoplasty), and 144 excisional biopsies. Pre-operative biopsies included invasive cancer (n=297, 56.14%), DCIS (n=69, 13.04%), and B2/B3/B4 pathologies (n=163, 30.81%). Median time from Magseed insertion to surgery was 7 days, with 36 placed on the day of surgery, and the maximum duration was a Magseed inserted 173 days pre-operatively. Mean duration of surgery (skin incision to specimen out) was 18 minutes. Median specimen weight was 20.8g. Re-excision rate for positive or close margins was 19.8% (n=72), and the migration rate (>10mm) was 2.9% (n=16). There were no major complications relating to the Magseed recorded. Conclusion: Magseed localisation in breast surgery is a safe and reliable technique. Its ability to be inserted any time before surgery, ease of operator use and low migration rates has resulted in our unit transitioning to using it as the primary localisation technique, replacing ROLLIS which had constraints in COVID-related surgical delays and radiation issues.

Background: Axillary surgery in breast cancer has undergone progressive de-escalation. The recent Sentinel Node versus Observation after Axillary Ultrasound (SOUND) and Intergroup Sentinel Mamma (INSEMA) trials have demonstrated that sentinel lymph node biopsy may be safely omitted in selected patients. This study aimed to evaluate the clinical relevance and applicability of the SOUND and INSEMA trial criteria within the Australian and New Zealand context. Methods: Prospectively maintained data from the BreastSurgANZ Quality Audit (BQA) were analysed. Patients with invasive breast cancer treated between 2019 and 2023 who met the inclusion criteria of the SOUND and INSEMA trials were identified. Results: A total of 69,122 patients were treated for invasive breast cancer over the five-year study period, of whom 30,822 fulfilled the inclusion criteria of either the SOUND or INSEMA trials. Overall, 34.0% and 44.6% of patients with invasive breast cancer in Australia and New Zealand met the SOUND and INSEMA eligibility criteria, respectively. Tumour characteristics of BQA patients were more closely aligned with those reported in the SOUND trial than the INSEMA trial, although key differences in other clinicopathological features were observed. Conclusions: The SOUND and INSEMA trial criteria are applicable to over one-third of patients with invasive breast cancer in Australia and New Zealand, supporting the potential relevance of axillary surgery de-escalation in this population.

Tumour-promoting stromal changes is an early event in breast cancer. They have been proposed as therapeutic and diagnostic targets but with minimal translational success. This project aims at evaluating “cystine rich with EGF like domains 2” (CRELD2) – a tumour cells’ secreted protein that mediates breast cancer-tumour-stroma crosstalk, as the first in class biomarker of the early invasive breast cancer. This is a prospective, correlational study. 58 breast cancer patients were recruited from the RAH Breast Department. Intra-operative tumour core biopsy tissues used to assess CRELD2 levels. This data was analysed against the patient’s tumour and disease phenotypes. Ten patients with high levels of detectable CRELD2 and 10 patients with low/no CRELD2 were used for further stromal assessment. This subset of patients’ tumour slides was stained for stromal components associated with cancer-associated fibroblasts. The fluorescence signal was measured using intensity density in the average of 10 region of interest sections per patient. Results: This pilot study had shown that the presence of intra-tumoural CRELD2 levels was strongly correlated with known prognostic characteristics such as: T stage, lymph node metastasis (p value <0.05) and tumour grade (p=0.07). All stromal characteristics tested were increased in patients with high intra-tumoral CRELD2 levels compared to patients who had no/low CRELD2 levels. Conclusion: Our findings show the utility of CRELD2 as a novel biomarker of early invasion of breast cancer. Despite the many prognostic factors known at the diagnostic stage size of tumour and grade of tumour, these still do not accurately predict for lymph node metastasis. This pilot study has shown CRELD2 as an independent predictive marker of local metastasis to lymph nodes. Alterations in breast cancer stroma is also shown in this cohort and supports the functional action and pro-tumorigenic property of CRELD2 that should be considered for future targeted therapy in breast cancer.

Purpose The surgical management of breast cancer has evolved with expectations for improved surgical, functional, aesthetic and oncological outcomes. This study aimed to describe these trends over time in breast cancer management in a large metropolitan centre. Method This retrospective cohort study assessed patients with breast cancer treated by four surgeons at the Westmead Breast Cancer Institute from 1 Jan 2015 to 31 Dec 2024. Data was extracted from surgeons’ logbooks and chart review, for procedures in both public and private hospitals. Linear regression models were used to identify statistical significance of trends over time. Results A total of 5709 patients were included. The rate of breast conserving surgery (BCS) increased progressively from 61% to 79%, while mastectomies decreased from 39% to 21% (p<0.01). Simultaneously, the proportion of neoadjuvant systemic therapy increased from 6% to 27% (p<0.01). The use of localisation for non-palpable disease doubled from 29% to 62% (p<0.01). Oncoplastic breast surgical techniques increased from 37% to 56% of BCS cases, with local perforator flap and immediate lipofilling increasing from 5% in 2018 to 22% in 2024 (p<0.01). The rate of immediate post-mastectomy reconstruction remained stable at around 45%; the proportion of implant-based vs autologous reconstruction changed from 95% vs 5% in 2020 to 51% vs 49% in 2024 (p<0.01). Symmetrisation of the contralateral breast doubled from 11% to 23% (p<0.01). The proportion of axillary lymph node dissection (ALND) decreased from 16% to 6%, whereas targeted axillary dissection (TAD) increased from 0% to 10% (p<0.01). Of patients with upfront TAD, 36% proceeded to cALND. Lymphovascular anastomosis was introduced in 2024. Conclusion Breast surgery has moved away from mastectomies and towards breast conservation, with de-escalation of axillary surgery and improvement in cosmetic outcomes. These changes reflect advances in oncoplastic breast surgery and use of neoadjuvant therapy.

Purpose: Following a breast cancer diagnosis, patients face complex, time‑sensitive decisions guided by multidisciplinary team (MDT) input, which can cause confusion and decisional conflict. Standalone large language models (LLMs) are unreliable for clinical decision support due to hallucinations and “black‑box” reasoning. This study aimed to develop and validate a hybrid AI chatbot that provides personalised, guideline‑based treatment summaries in plain language. Methodology: A bespoke web‑based hybrid AI chatbot was developed (in collaboration with IT and Computer Scientists), combining deterministic NCCN‑aligned decision logic—implemented through a structured decision matrix mapping clinical variables to recommended options—with a conversational LLM interface (Claude Sonnet 4.5). This architecture ensures guideline‑compliant outputs delivered with a patient‑friendly tone. Validation is underway using 150 retrospective cases through a two‑stage concordance analysis comparing chatbot outputs with (1) initial clinical decisions and (2) postoperative histopathology‑informed MDT plans (concordance rates and Cohen’s kappa). After validation, supervised machine learning (ML) optimisation using 1000 retrospective MDT cases will enhance classifier accuracy, triage performance, and alignment with real‑world decision patterns. Results: Development of the functional chatbot is complete. Early testing shows consistent generation of personalised and guideline‑concordant summaries. Ongoing validation is assessing concordance, sensitivity, specificity, and factors contributing to discrepancies. Conclusion: A novel, explainable hybrid AI chatbot for breast cancer decision support has been developed. Validation and ML optimisation are in progress, with final results and optimised model planned for presentation by April 2026. This approach aims to deliver a transparent, reproducible, and patient‑centred AI tool with future plans for real-world pilot trial.

Introduction Breast conserving surgery has good oncological outcomes for selected breast cancers and may result in superior cosmetic outcomes. Traditionally used for smaller tumours, utilisation has expanded with the development of oncoplastic techniques such as local perforator flaps. This study describes the experience of introducing an intercostal perforator flap service at a tertiary hospital in Western Australia. Methods This case series used retrospectively collected data on patients who underwent breast conserving surgery with intercostal perforator flap reconstruction at Sir Charles Gairdner Hospital between May 2023 and September 2025. Data collected included flap type, tumour characteristics, complications, and re-operation rates. Pre-operative tumour size on imaging was compared with final histological size using Student’s t-test. Results 28 consecutive patients were included, with all reconstructions performed immediately. The mean operative time was 122 minutes (range 60–245). Flaps were most commonly based on the lateral intercostal perforator, with 28.6% of cases utilising dual-perforator supply. The median specimen weight was 38.6 g (range 18.6–250 g). Mean tumour size was 28.3 mm on both pre-operative imaging and final histology, with no statistically significant difference. Four patients underwent surgery for ductal carcinoma in situ alone, while the remainder had invasive cancer ranging from T1–T3 tumours.18 patients underwent concurrent sentinel lymph node biopsy while 6 had axillary clearance. Margin re-excision was required in four patients (14.3%) and post-operative infection occurred in six patients (21.4%), half of whom required procedural intervention. There were no post-operative haematomas or complete flap losses. Conclusion The introduction of an intercostal perforator flap service enabled breast conserving surgery for a broad range of tumour sizes and pathologies with acceptable operative times and key performance indicators.

Purpose: Immediate autologous breast reconstruction has many advantages but presents logistical challenges. Establishing a coordinated, combined model of care delivery has been shown to increase breast unit productivity at minimal cost. This study sought to validate this novel intervention by evaluating clinical and patient-reported outcomes following its implementation. Methods: A retrospective cohort study was conducted involving patients who underwent immediate autologous breast reconstruction at a Western Australian tertiary centre over five years. The BREAST-Q v2.0 Reconstruction Module was used to collect patient-reported outcome data. Clinicopathological data were analysed using SPSS v.27. Results: 105 patients were included, with a mean age of 48.8 (SD ± 9.5) and mean BMI of 27.0 (SD ± 4.3). The vast majority of reconstructions (84.8%) were DIEP flaps. Overall 90-day complication rate was 37.1%, with rates of unplanned re-operation and flap failure of 10.5% and 1.0%, respectively. The most frequent complication was wound dehiscence (8.6% breast, 17.1% donor site). BREAST-Q response rate was 48.1%. Overall breast satisfaction was high, with 84% of patients answering ‘satisfied’ more than ‘dissatisfied’. Sexual well-being and psychosocial well-being scores were lower than in prior literature, though this may be influenced by a significant number of patients still awaiting aesthetic revision. Patient-reported outcomes were poorer among those without a cancer diagnosis, highlighting the importance of a robust risk-benefit discussion prior to mastectomy and reconstruction. Conclusion: The findings of this follow-up study showed serious complication rates and breast satisfaction on par with literature benchmarks. This further supports its application in other healthcare settings.