Evaluation of pretreatment ADC values as predictors of treatment response to neoadjuvant chemotherapy in patients with breast cancer - a multicenter study

Background Magnetic resonance imaging (MRI) can be used to diagnose breast cancer. Diffusion weighted imaging (DWI) and the apparent diffusion coefficient (ADC) can reflect tumor microstructure in a non-invasive manner. The correct prediction of response of neoadjuvant chemotherapy (NAC) is crucial for clinical routine. Our aim was to compare ADC values between patients with pathological complete response (pCR) and non-responders based upon a multi-center design to improve the correct patient selection, which patient would more benefit from NAC and which patient would not. Methods For this study, data from 4 centers (from Japan, Brazil, Spain and United Kingdom) were retrospectively acquired. The time period was overall 2003–2019. The patient sample comprises 250 patients (all female; median age, 50.5). In every case, pretreatment breast MRI with DWI was performed. pCR was assessed by experienced pathologists in every center using the surgical specimen in the clinical routine work up. pCR was defined as no residual invasive disease in either breast or axillary lymph nodes after NAC. ADC values between the group with pCR and those with no pCR were compared using the Mann–Whitney U test (two-group comparisons). Univariable and multivariabe logistic regression analysis was performed to predict pCR status. Results Overall, 83 patients (33.2%) achieved pCR. The ADC values of the patient group with pCR were lower compared with patients without pCR (0.98 ± 0.23 × 10 − 3 mm 2 /s versus 1.07 ± 0.24 × 10 − 3 mm 2 /s, p = 0.02). The ADC value achieved an odds ratio of 4.65 (95% CI 1.40–15.49) in univariable analysis and of 3.0 (95% CI 0.85–10.63) in multivariable analysis (overall sample) to be associated with pCR status. The odds ratios differed in the subgroup analyses in accordance with the molecular subtype. Conclusions The pretreatment ADC-value is associated with pathological complete response after NAC in breast cancer patients. This could aid in clinical routine to reduce treatment toxicity for patients, who would not benefit from NAC. However, this must be tested in further studies, as the overlap of the ADC values in both groups is too high for clinical prediction.