Mechanistic modeling of signaling pathways mediating patient-specific response to therapy can help to unveil resistance mecha-nisms and improve therapeutic strategies. Yet, creating suchmodels for patients, in particular for solid malignancies, is chal-lenging. A major hurdle to build these models is the limited mate-rial available that precludes the generation of large-scaleperturbation data. Here, we present an approach that couplesex vivohigh-throughput screenings of cancer biopsies usingmicrofluidics with logic-based modeling to generate patient-specific dynamic models of extrinsic and intrinsic apoptosis signal-ing pathways. We used the resulting models to investigate hetero-geneity in pancreatic cancer patients, showing dissimilaritiesespecially in the PI3K-Akt pathway. Variation in model parametersreflected well the different tumor stages. Finally, we used ourdynamic models to efficaciously predict new personalized combi-natorial treatments. Our results suggest that our combination ofmicrofluidic experiments and mathematical model can be a noveltool toward cancer precision medicine.