Factors associated with axillary conversion after neoadjuvant chemotherapy (NAT) in initially node positive breast cancer patients - a transSENTINA analysis
Kolberg H.-C.1, Kühn T.2, Krajewska M.3, Bauerfeind I.4, Fehm T.N.5, Fleige B.6, Helms G.7, Lebeau A.8, Stäbler A.7, Loibl S.9, Untch M.6, Kolberg-Liedtke C.3
1Marienhospital Bottrop, Bottrop, Deutschland, 2Klinikum Esslingen, Esslingen, Deutschland, 3Charité - Universitätsmedizin Berlin, Berlin, Deutschland, 4Klinikum Landshut, Landshut, Deutschland, 5Universitätsklinikum Düsseldorf, Düsseldorf, Deutschland, 6Helios Klinikum Berlin Buch, Berlin, Deutschland, 7Universitätsklinikum Tübingen, Tübingen, Deutschland, 8Universitätsklinikum Hamburg - Eppendorf, Hamburg, Deutschland, 9German Breast Group, Neu Isenburg, Deutschland

Background: Selection criteria for the strategy of omission of axillary surgery in selected patients after NAT have to include the probability of conversion from cN1 to ycN0. We analyzed the association of clinical/pathological parameters and axillary conversion in arms C and D of the SENTINA trial.
Methods: Patients in arms C/D of the SENTINA trial presented with clinically positive nodes before NAT. Based on their response to NAT they were assigned to arm C (ycN0) or arm D (ycN+). The association between clinical/pathological parameters and axillary conversion after NAT was analyzed by univariate logistic regression.
Results: Of the 892 patients in arms C and D of the SENTINA trial 716 were evaluable. 593 patients converted to ycN0 (arm C), 123 patients still had involved lymph nodes after NAT (ycN+) (arm D). Arms C and D were compared regarding tumor diameter by ultrasound before and after NAT, grading, multifocality, ER, PR, HER2, pathological complete remission in the breast (breast pCR), morphology, lymphovascular invasion (LVI) and hemangiosis. Small tumor diameter after NAT (p=0.0038), breast pCR (p=0.0001) and lack of LVI (p=0.0009) were positively associated with axillary conversion.
Conclusion: Because of the small patient number in arm D, we could not identify an association between tumor biology and axillary conversion. Clinical response of the primary tumor and breast pCR were positively associated with conversion from cN1 to ycN0. These results justify including patients with clinical and pathological response of the primary tumor in trials investigating de-escalation of axillary surgery after NAT.