Abstract PO3-23-07: Evaluation of the role of axillary ultrasound to estimate axillary nodal burden to guide de-escalation of upfront axillary surgery in node positive early breast cancer patients

Abstract Background: Axillary surgery is now routinely de-escalated in clinically node negative breast cancer patients with limited axillary nodal burden (up to 2 involved sentinel nodes) based on the results of ACOSOG-Z0011 and AMAROS trials. However, patients with proven axillary metastasis following on ultrasound (AUS) guided core needle biopsy (CNB), undergoing primary surgery are still subjected to an axillary lymph node dissection (ALND) irrespective of the magnitude of the axillary nodal burden. If we could predict axillary nodal burden with a low false negative rate (FNR) for the presence of extensive axillary nodal involvement, we can then hypothesize that axillary surgery could be safely de-escalated in this cohort of patients. We investigated whether pre-operative AUS suggesting limited axillary burden is sensitive enough to rule out gross axillary disease burden. Methods: A single institution retrospective study. We undertook an electronic chart review of all T1 to T3 patients undergoing upfront ALND with either mastectomy or breast conserving surgery between January 2019 and June 2023. Patients with palpable axillary nodes, locally advanced breast cancer or a completion ALND after positive sentinel lymph node biopsy (SLNB) were excluded from the study. Clinical and pathological characteristics associated with nodal burden of disease were evaluated. Limited axillary burden was defined as 2 involved nodes with ≥3 taken as gross nodal burden for primary analyses. Sensitivity analyses were carried out with limited axillary burden defined as 3 involved nodes (N1) and ≥4 (N2) taken as gross nodal burden. Subgroup analyses in T1 or T2 tumors were carried out as well. The probabilities of discordant events, defined as limited burden on AUS-CNB but gross burden on final histopathology evaluation (HPE), assuming 10% FNR, were estimated using inverse cumulative binomial distribution function. Results: Data on the exact number of suspicious nodes on AUS were missing for 2 patients, therefore, the final study cohort included 39 patients. The correlation between AUS nodal burden and that on HPE was moderate (Spearman’s rho = 0.48; p = 0.0018). The sensitivity and specificity of AUS in identifying gross nodal burden were 61% and 81% respectively. Observation of 9 discordant events in 22 patients with low burden on AUS ruled out an FNR of 10% with 99.99% certainty. The sensitivity and specificity in T1/T2 subgroup were 60% and 75% respectively. The sensitivity analyses classifying N1 as limited and N2 as gross axillary burden showed the sensitivity and specificity of 35% and 82% respectively and an FNR of 10% was ruled out with 99.996% certainty. Conclusion: Pre-operative AUS alone lacks the sensitivity and specificity to accurately predict the true axillary nodal burden to safely guide de-escalation of axillary surgery. AUS followed by targeted axillary dissection (TAD) in conjunction with a dual technique SLNB may provide the desired sensitivity and specificity to guide de-escalation of axillary surgery. A prospective evaluation of this approach is merited. Given the limited number of patients that could be enrolled in such prospective study from individual institutions, an international collaborative effort is needed to address this important research question. Citation Format: Nidhi Garg, Mangesh Thorat, Hisham Hamed, Ashutosh Kothari. Evaluation of the role of axillary ultrasound to estimate axillary nodal burden to guide de-escalation of upfront axillary surgery in node positive early breast cancer patients abstract. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-23-07.