Changes in the findings of dynamic MRI by preoperative CAF chemotherapy for patients with breast cancer of stage II and III: Pathologic correlation

Preoperative neoadjuvant chemotherapy is essential for treatment of patient s with breast cancer who have a large tumor mass and/or regional lymph node involvement, in terms of both tumor shrinkage and further improvement of t he survival rate. In order to safely perform breast-conservation treatment for these patients, a detailed diagnostic procedure for precisely evaluatin g the therapeutic response is needed. Dynamic magnetic resonance imaging (M RI) is thought to be important in the evaluation of responses to neoadjuvan t therapy in patients with considerably large tumors, however, few studies have detailed the changes, as depicted by dynamic MRI, that can be expected with neo-adjuvant chemotherapy. The purpose of this study was to document the changes that occur in response to neoadjuvant chemotherapy and to corre late them with the pathological findings observed in the surgical specimen. The study was performed at Kochi Medical School Hospital from 1995 to 1998 . The series consisted of 31 patients with stage II and III breast cancer. Prior to and after 1-5 courses of neoadjuvant chemotherapy, dynamic MRI exa minations were performed. Eight of the time-intensity curves for the 10 gra de la tumors flattened during neoadjuvant chemotherapy, while two remained the same. Six of the curves flattened for the 14 grade 1b tumors, 7 remaine d the same, and one spiked. And for the seven grade ? tumors, two of the cu rves flattened and five remained the same (p=0.0340). In the five grade 1 t umors, the mean after/before normalized peak signal intensity ratio was 0.4 2+/-0.22. In the 18 grade 2 and 8 grade 3 tumors, the mean normalized signa l intensity ratios were 0.59+/-0.28, 0.88+/-0.10, respectively (p<0.05). In the 15 tumors that showed shrinkage of the linear enhancement during neo-a djuvant chemotherapy, 10 had no remarkable intraductal spreading and 9 had a negative surgical margin. In the 16 tumors that had no shrinkage of the l inear enhancement during chemotherapy, 13 had remarkable intraductal spread ing and 12 had a positive surgical margin (p<0.05). It is concluded that dy namic MRI is a valuable tool for determining tumor response and predicting a positive surgical margin. Breast-conservation treatment can be performed for these patients by meticulous assessment using such detailed diagnostic procedures after local tumor control by combined chemo-therapy with high do se-intensity.