332PD - Prediction of axillary lymph node response to neoadjuvant chemotherapy (NAC) by changes in maximum standard uptake values (SUVmax) with 18FDG-PET (...

Date 28 September 2014
Event ESMO 2014
Session Metastatic and locally advanced breast cancer: Facing with heterogeneity and endpoints in clinical trials
Topics Breast Cancer, Locally Advanced
Imaging, Diagnosis and Staging
Pathology/Molecular Biology
Presenter Muhammad Alamgeer
Citation Annals of Oncology (2014) 25 (suppl_4): iv110-iv115. 10.1093/annonc/mdu328
Authors M. Alamgeer1, M. White2, J. Stuckey3, M. Harris2, B. Kumar4, S. Hart1, J. Fox5, D.N. Watkins6, V. Ganju2
  • 1Oncology, Monash Health, 3165 - East Bentleigh/AU
  • 2Medical Oncology, Monash Medical Centre, East Bentleigh/AU
  • 3Diagnostic Imaging, Monash Medical Centre, East Bentleigh/AU
  • 4Department Of Pathology, Monash Health, 3165 - East Bentleigh/AU
  • 5Breast Oncology Centre, Monash Health, 3165 - East Bentleigh/AU
  • 6Centre For Cancer Control, MIMR-PHI Institute of Medical Research, Clayton/AU

Abstract

Aim

The aim of this study was to investigate the effect of NAC (Fluorouracil, epirubicin, cyclophosphamide (FEC100) or docetaxel) in predicting pathological complete response in axillary nodes (pCRA) based on changes in SUVmax on PET in the primary tumour in locally advanced breast cancer (LABC).

Methods

Women with node positive LABC were randomized to receive 4 cycles of FEC100 followed by four cycles of docetaxel (Arm A), or the regimen in reverse order (Arm B). FDG-PET studies were performed at baseline, after 4 and 8 cycles of NAC. pCRA to NAC was assessed in surgical specimens. Receiver operating characteristic analysis (ROC) was performed to determine a cut-off value Δ% SUV 1-2 and SUV 1-3 for prediction of nodal status after NAC. Association between SUVmax changes and pCRA was analysed by Mann Whitney U test.

Results

One hundred women with node positive LABC were enrolled into the study, 52 (%) were treated on Arm A and 48 (%) on Arm B. Ninty five (95%) had invasive ductal cancers, 64(%) were ER + , 32(%) were HER2 + , while 23(%) were triple negative subtype. Baseline characteristics and PET SUVmax values were similar in two Arms. Baseline PET SUVmax correlated with tumour grade (P = 0.003), triple negative phenotype (P = 0.014) and Ki67 index (P = 0.005). All women had axillary lymph node dissection with fifty one (51%) undergoing breast conserving surgery. pCRA was seen in 46 (%) women. Overall there was no significant difference in pCRA rates between the two drug regimens. PET SUV reduction after four cycles of FEC was a strong predictor of pCRA. In Arm A, pCRA was associated with median PET SUVmax reduction of 81% after 4 cycles of FEC, in contrast to only 39% reduction after 4 cycles of docetaxel in Arm B (P < 0.001). The sensitivity, specificity, positive and negative predictive values of PET SUVmax reduction at the treatment midpoint (4 cycles) were 92%, 56%, 69% and 87%, respectively for Arm A, while 40%, 57%, 55%, and 46% for Arm B.

Conclusions

PET SUV response in the primary tumour, after 4 cycles of FEC100 accurately predicts the axillary status. By contrast assessment of PET response with docetaxel resulted in higher rates of false positive and negative. This methodology may allow for selection of patients for less aggressive axillary surgery after NAC.

Disclosure

All authors have declared no conflicts of interest.