Value of early metabolic response for predicting axillary pathological complete response during neoadjuvant systemic therapy in triple negative and HER2-amplified breast cancers: impact of tumor subtypes.

Background: In the era of therapeutic de-escalation, the opportunity to move from systematic axillary lymph node dissection (ALND) to sentinel lymph node biopsy in axillary node-positive breast cancer patients after neoadjuvant systemic therapy (NST) is currently considered. The purpose of this study was to identify FDG-PET parameters associated with axillary pathological complete response (pRAx) in the most proliferative tumor subtypes, eg Triple Negative (TN) and HER2-amplified.

Methods: Patients with newly-diagnosed TN or HER2-amplified breast cancer, with pathologically-proven axillary node metastasis, no distant metastasis and indication of NST were prospectively included from September 2017 to December 2021. Sequential FDG-PET/CT scans were performed at baseline and after one cycle of NST. Metabolic parameters at baseline and their changes (Delta in %) of axillary nodes were assessed: SUVmax, SUVratio (SUVmax/SUVmax liver), SUVpeak, SUVmean, TLG and MTV. Logistic regressions with ROC curves were used to determine parameters associated with pRAx.

Results: Sixty-one patients (24 TN, 19 ER-negative/HER2-amplified and 18 ER-positive/HER2-amplified) were recruited. Median value of Axillary SUVmax at baseline were 7.9, 7.0 and 5.2 and Delta Axillary SUVmax were -62%,-60% and -47% in these 3 subgroups, respectively. In univariate model, in the whole population, Delta Axillary SUVmax showed the greatest AUC for prediction of pRAx of 0.72 (95%CI: 0.59-0.85), whereas AUC of Axillary SUVmax at baseline was not statistically significant (AUC = 0.6 (95%CI: 0.46-0.74)). Specificity, sensitivity, PPV and NPV of Delta Axillary SUVmax were 96%, 49%, 94% and 58% respectively for predicting pRAx with a threshold of -68.7%. Odd Ratio associated with Delta Axillary SUVmax < -68.7% compared to ≥ -68.7% was 24.0 (95%CI: 2.9-194). In multivariate model, adjusted on tumor subtypes, Delta Axillary SUVmax was still significantly associated with pRAx (OR = 20.7 (95%IC: 2.5-172). AUCs adjusted on the tumor subtype were not significantly modified compared to univariate model (p = 0.45 compared to unadjusted AUC) suggesting that thresholds were not significantly different in each tumor subtype.

Conclusions: Delta Axillary SUVmax seems to be the most relevant metabolic parameter to predict an axillary pathological complete response and early metabolic response could be a valuable tool for selecting patients eligible for axillary surgical de-escalation after NST, regardless tumor subtypes.