Abstract Physiologically relevant in vitro tumor models are crucial in any research setting from preclinical drug development to functional precision oncology. Patient-derived 3D cell culture models (PD3D®) are validated cancer models which recapitulate the biology of the original tumor tissue. PD3D® can be used to model intratumoral heterogeneity in medium and high throughput screens. Using a reverse clinical engineering approach, PD3D® models are already applied in personalized oncology to identify treatment for an individual patient. We successfully established a living biobank of more than 500 PD3D® models, spanning from common cancers like colorectal, breast (incl. TNBC), non-small cell lung and pancreatic carcinoma, to orphan indications including various subtypes of sarcomas. Not surprisingly, PD3D® model morphology and culture requirements differ between tumor entities. Treating PD3D® models with standard of care drugs in a semi-automated high-throughput assay, we observed heterogeneity in drug sensitivity between models of the same cancer type, recapitulating clinical response of patients. Combining drug sensitivity profiles with genomic and proteomic data of the same PD3D® models, we successfully identified a biomarker for predicting chemosensitivity towards MEK-targeting drugs. For application of PD3D® in truly personalized oncology, we developed a protocol that allows us to generate a PD3D® culture and perform a drug sensitivity assay for an individual patient within a therapy-relevant timeframe. Within 38 days we identified a systemic therapy for a young patient with a relapsed metastasized synovial sarcoma. After surgery and initiation of recommended chemotherapy, the patient is now under remission. To further expand the scope of PD3D® for therapy response prediction, we established a method for irradiation of PD3D® cultures with different dosages of photon and proton radiation. We observed significant, model specific differences in radiosensitivity between PD3D® sarcoma models which also reflect clinically heterogenous responses. Additionally, PD3D® models can be employed for a tumor organoid-on-chip platform (TumOC), delivering real-time information on physiological parameters like drug response-mediated oxygen consumption in a microfluidic system. In conclusion, PD3D® models act as jacks-of-all-trades in current and future cancer research, delivering robust and reliable data that not only is of academic value, but speeds up the drug development process, and are ready for prime time in functional precision oncology. Citation Format: Lena Wedeken, Samantha Forbrig, Katja Herrera-Glomm, Juergen Loskutov, Ulrike Pfohl, Irina Piven, Michael Poehle, Manuela J. Regenbrecht, Barbara Seller, Cynthia Yapto, Sabine Finkler, Larissa Ruhe, Quirin Graf Adelmann, Christoph Reinhard, Marie Flechner, Katja Uhlig, David Kaul, Siyer Roohani, Maya Niethard, Rica Sauer, Christian R. Regenbrecht. PD3D®models as jacks-of-all-trades for cancer research and therapy response prediction [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4235.