Risk Stratification and Surgical De-Escalation in T1 Invasive Lobular Carcinoma: A Retrospective Analysis of Preoperative Axillary Imaging

Background: Invasive lobular carcinoma (ILC) is a common histological subtype of breast cancer, yet preoperative axillary staging remains notoriously challenging due to its diffuse, infiltrative growth pattern.Given the increasing clinical shift toward axillary surgical de-escalation for low-volume disease, precise preoperative evaluation is crucial to guide surgical planning.Objective: This study aimed to determine the efficacy of preoperative axillary ultrasound and magnetic resonance imaging (MRI), alongside fine needle aspiration cytology (FNAC), in staging ILC and to evaluate the potential for safely omitting sentinel lymph node biopsy (SLNB) in a risk-stratified subgroup.Methods: A retrospective analysis was conducted on 167 patients diagnosed with operable primary ILC (T1-3, N0-2, M0) treated at a single healthcare trust between February 2020 and August 2024.Electronic care records were reviewed for preoperative imaging and FNAC results, which were then correlated with final postoperative pathology.For nodal classification, isolated tumour cells (ITCs) (pN0(i+)) were treated as node-negative, whereas micrometastases (pN1(mi)) and macrometastases were treated as node-positive.Axillary ultrasound was performed first, with FNAC performed where ultrasound findings required cytological assessment.A negative preoperative axillary assessment was defined as no suspicious nodes on ultrasound and, where FNAC was performed, negative cytology.Nodal status was categorised by burden, and sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for each imaging modality.Results: Of the 167 patients, 58 (34.7%) had node-positive disease.Axillary ultrasound demonstrated an overall sensitivity of 32.8% and a specificity of 100%, while MRI yielded a sensitivity of 29.6% and a specificity of 98.0%.Ultrasound sensitivity varied numerically across tumour-size categories and nodal burden, falling to 0.0% in T1 tumours compared to 40.9% in T2 tumours.Crucially, the prevalence of a high nodal burden (≥3 positive nodes) was extremely rare in T1 tumours (2.4%), in stark contrast to T2 (11.8%) and T3 (45.2%)tumours.The proportion of high nodal burden was low in both age groups.Conclusions: Neither ultrasound nor MRI can reliably rule out low-volume nodal disease in ILC due to the inherent biological characteristics of the tumour.However, both modalities provide excellent specificity.The high NPV of ultrasound alone in T1 tumours is driven by the fundamentally low prevalence of extensive axillary disease in this subgroup, rather than diagnostic utility.Given that high nodal burden is exceedingly rare in T1 tumours, the hypothesis is that SLNB could potentially be omitted in selected T1 ILC patients.These findings support a risk-stratified approach, suggesting that patients with T1 ILC and a negative preoperative axillary assessment (no suspicious nodes on ultrasound and negative cytology where FNAC was performed) could be considered for surgical de-escalation and omission of SLNB, pending prospective validation.