The study evaluated whether the analysis of the distribution of diffusion coefficient (ADC) values from MRI of the whole tumor can preoperatively predict lymph node metastases in ductal adenocarcinoma of the pancreas. 53 patients with a histologically confirmed tumor were included in the retrospective analysis, of which 29 with metastases in the lymph nodes and 24 without metastases. ADC maps were generated from diffusion-weighted MRIs and histogram parameters were calculated for the entire tumor volume. Most ADC histogram parameters, except coefficient of variation and kurtosis, were significantly different between groups (p < 0.05), and first-order ADC values were significantly lower in patients with metastases. Baseline clinical data (age, sex, symptoms, CA19-9) and conventional MRI parameters (tumor size and volume) did not differ significantly between groups. The multiparametric model based on selected ADC metrics achieved an area under the ROC curve (AUC) of 0.865, a sensitivity of 86.2% and a specificity of 75.0%. The authors conclude that whole-tumor histogram analysis of ADC represents a non-invasive quantitative tool for preoperative prediction of lymph node metastases in this type of pancreatic cancer and that a multiparametric model has better diagnostic performance than individual parameters alone.