Korrelation zwischen Biopsaten und postinterventionellen Resektaten beim Mammakarzinom: eine retrospektive Kohortenanalyse

This is a single-center, retrospective analysis evaluating the concordance of histological type, nuclear grade, hormone receptors, HER2/neu status, Ki67, lymphangiosis carcinomatosa, haemangiosis carcinomatosa and perineural invasion between biopsies and surgical specimens. We identified 1280 breast tumors form 01.01.2007 to 31.12.2016 from our database at the breast cancer center ‘Sana Klinikum Offenbach’. Thus, we excluded 30 cases with local recurrences. The analysis was performed in 1002 breast tumors with upfront surgery (K1.1). In total 248 tumors were assigned to neo-adjuvant systemic treatment, resulting in 60 with a pCR (in 1 case with cCR, the patient refused surgery). A pCR as best response was identified in 187 cases (K1.2.2). In cohort K1.1 we identified the following discrepancies: 15.6%, 28.4%, 2%, 10%, 6%, 20.3%, 3.6%, 15.4%, 2.5%. In cohort K1.2.2, however we found discrepancies 8.7%, 8.3%, 27.1%, 11.4%, 52.1%, 4.3%, 22.7% and 1.1%, respectively. We found different changes depending of the timing of systemic therapy. Significant differences were detected in the discrepancy of the following factors without NACT: histological type, nuclear grade, progesterone receptors, Ki67 and haemangiosis carcinomatosa, and with NACT: estrogen receptors, progesterone receptors, HER2/neu status, Ki67 and haemangiosis carcinomatosa. There was no difference in cohort K1.1 for patients who received more or less than three core biopsies. The discrepancy of nuclear grading was higher in the group of NST tumors, the discrepancy of estrogen receptors was higher in the group of cT4 tumors as well as when CNB was taken. The highest discrepancy of Ki67 was identified in the group of lobular tumors. We also evaluated the molecular subtype based on CNB and surgical specimen in both cohorts. The discrepancy in K1.1 was 18% for luminal A, 33% for luminal B HER2/neu-negative, 27% for luminal B HER2/neu-positive, 19% for HER2/neu-positive und 10% for basal-like tumors. In K1.2.2 it was 31% for luminal A, 49% for luminal B HER2/neu-negative, 39% for luminal B HER2/neu- positive, 23% for HER2/neu-positive und 9% for the basal-like tumors. Our Study indicates that despite the tumor heterogeneity, CNB and VSB can provide reliable Information on the estrogen receptors, HER2/neu status, lymphangiosis carcinomatosa and perineural invasion. We suggest a more cautious approach regarding tumor type, nuclear grade, progesterone receptors, Ki67 and haemangiosis carcinomatosa and repeating testing in surgical specimen, especially when a discrepancy might lead to a change in treatment. We suggest that not only fixational and observational issues, but also the heterogeneity of breast cancer can be the leading cause of those discrepancies. Our data suggests that NACT might modulate prognostic factors, as well as the molecular subtype. This finding might be hypothesis generating if this modulation or shifting might give rise to some new thoughts about post-neoadjuvant therapy approaches, like currently investigated in the context of clinical trials (e.g. I-SPY Trial, ADAPT, ADAPT-Cycle). Current therapy recommendations speak for endocrine therapy regardless of the post- neoadjuvant HR status. In case of HER2/neu-positive breast cancer, the composition of post-neoadjuvant anti-HER2 therapy solely based on the remission status (pCR vs. non-pCR); however, in pCR, some data might suggest do de-escalate a dual blockade towards trastuzumab monotherapy. In the case of non-pCR, T-DM1 for the completion of 1 year is recommended. Remarkably in non-pCR with a negativ shift (HER2/neu-positiv to negativ) the post-neoadjuvant anti-HER2/neu treatment will be based on the data derived from the biopsy. At present experimental data are suggesting that this treatment related changes are transient in nature, not qualifying for a definite change of treatment. In summary, the unmet clinical need remains if the neo-adjuvant systemic therapy might identify a cellular selection for a different treatment approach.