Pathological response to neoadjuvant chemotherapy for inflammatory breast cancer (IBC). Critical evaluation of Chevallier’s and Sataloff’s classifications

10666 Background: The aim of this study was to compare the two most frequently used pathological classifications to assess the response to chemotherapy (CT) in IBC. Methods: The pathological characteristics of 85 inflammatory breast cancers were reviewed. 52 patients (pts) had a post-CT mastectomy , 49 a tumor biopsy before CT and 16 both. Tumor types, grade, hormonal and Her status were evaluated before and after CT, when available. Results: Among 49 pts with initial biopsy, 1 had a high grade intraductular carcinoma and 48 an invasive carcinoma (42 ductular, 5 lobular, 1 mucinous). 27 (56%) invasive tumors were grade 3 and 21 (44%) grade 2, none was grade 1. Median value of mitotic index (Ki 67) was 23%. Hormone receptors (HR) were evaluable for 46 pts; 28 (61%) were negative. 13 (29%) out of 45 evaluable pts were Her2 3+. The pathological response was evaluated according to the Chevallier et Sataloff classifications in the 52 pts with post CT- mastectomy. According to Chevallier there were no grade 1(no microscopic invasive or in situ carcinoma) or 2 (microscopic in situ carcinoma without invasive carcinoma or axillary lymph node metastases), 44 (85%) were grade 3 (invasive carcinoma with fibrosis or sclerosis) and 8 grade 4 (no modification of initial tumor). There was no pathological complete response. According to Sataloff there were 4 grade TA (complete or nearly complete response), 15 (29%) TB (therapeutic response over 50%), 23 (44%) TC (therapeutic response above 50%), 10 TD (no evident therapeutic effect). There were 4 pathological complete responses out of 52 ( 8%). 16 pts had initial biopsy and post chemotherapy mastectomy. According to the OMS clinical classification there were 5 CR, 37 PR, 10 ST. Among the 4 pathological CR according to Sataloff (TA), 2 were CR and 2 PR. Conclusions: In our experience, there was a very poor correlation between the two pathological classifications and between pathological and clinical classifications. A clinical residual tumor can be due to fibrosis only. The classification used for the determination of the pathological response should always be given in clinical studies. No significant financial relationships to disclose.