Integrated 18F-FDG PET/MRI in breast cancer: early prediction of response to neoadjuvant chemotherapy

Purpose To explore whether integrated F-18-FDG PET/MRI can be used to predict pathological response to neoadjuvant chemotherapy (NAC) in patients with breast cancer. Methods Between November 2014 and April 2016, 26 patients with breast cancer who had received NAC and subsequent surgery were prospectively enrolled. Each patient underwent F-18-FDG PET/MRI examination before and after the first cycle of NAC. Qualitative MRI parameters, including morphological descriptors and the presence of peritumoral oedema were assessed. Quantitatively, PET parameters, including maximum standardized uptake value, metabolic tumour volume and total lesion glycolysis (TLG), and MRI parameters, including washout proportion and signal enhancement ratio (SER), were measured. The performance of the imaging parameters singly and in combination in predicting a pathological incomplete response (non-pCR) was assessed. Results Of the 26 patients, 7 (26.9%) exhibited a pathological complete response (pCR), and 19 (73.1%) exhibited a non-pCR. No significant differences were found between the pCR and non-pCR groups in the qualitative MRI parameters. The mean percentage reductions in TLG(30%) on PET and SER on MRI were significantly greater in the pCR group than in the non-pCR group (TLG30% -64.8 +/- 15.5% vs. -25.4 +/- 48.7%, P = 0.005; SER -34.6 +/- 19.7% vs. -8.7 +/- 29.0%, P = 0.040). The area under the receiver operating characteristic curve for the percentage change in TLG30% (0.789, 95% CI 0.614 to 0.965) was similar to that for the percentage change in SER (0.789, 95% CI 0.552 to 1.000; P = 1.000). The specificity of TLG30% in predicting pCR) was 100% (7/7) and that of SER was 71.4% (5/7). The sensitivity of TLG30% in predicting non-pCR was 63.2% (12/19) and that of SER was 84.2% (16/19). When the combined TLG30% and SER criterion was applied, sensitivity was 100% (19/19), and specificity was 71.4% (5/7). Conclusion F-18-FDG PET/MRI can be used to predict non-pCR after the first cycle of NAC in patients with breast cancer and has the potential to improve sensitivity by the addition of MRI parameters to the PET parameters.