Histologic breast cancer extent after neoadjuvant chemotherapy: comparison with multidetector-row CT and dynamic MRI.

PURPOSE: To evaluate the efficacy of dynamic multidetector-row CT (MDCT) in assessing residual cancer extent after neoadjuvant chemotherapy (NAC), and to compare MDCT results with those derived from dynamic three-dimensional MRI using the volumetric interpolated breath-hold examination (VIBE) sequence. MATERIALS AND METHODS: MDCT before and after NAC was performed in 19 consecutive patients with breast cancer. MRI was also performed before surgery. The early phase of MDCT and MRI was started 60 sec after commencing contrast injection. The late phase was started at a 4-min delay from the injection. The injection rate was 3 mL/sec. The distribution pattern of contrast enhancement (CE) by CT before NAC was classified into two groups: replaced lesion (diffuse CE in whole quadrants) and non-replaced lesion (localized CE). RESULTS: Pathological complete response (pCR) was obtained in one case. In replaced lesions, accuracy for the detection of tumor extent with a deviation of less than 2 cm in length was 0% (0/7) with early-phase CT/MRI and 100% (7/7) with late-phase CT/MRI. In non-replaced lesions, accuracy was 55% (6/11) with early-phase CT/MRI and 82% (9/11) with late-phase CT/MRI. One case of ductal carcinoma in situ (DCIS) could be detected only with late phase MRI. CONCLUSION: Late-phase images obtained by MDCT and MRI may be accurate in the diagnosis of residual cancer extent after NAC. The tumor distribution determined by MDCT before NAC is thought to be useful in the evaluation of shrinkage pattern following NAC.