Abstract Introduction: Over the past 20 years, advances in precision oncology and cancer genetics have identified numerous actionable oncogenic drivers, leading to new drug development and clinical application. However, tumor cells can adapt and develop resistance mechanisms (MoR), and the mechanisms of action (MoA) of these drugs are only partially understood, highlighting the need for research focused on both MoR and MoA. Following the success of the MATCH-R clinical trial (NCT02517892, 2015-2022), the UNLOCK program was launched in 2022 to better understand the MoA and MoR of innovative drugs (phase 1 to approval) through integrated clinical and translational research. As part of this initiative, we developed preclinical models, including PDX models from patients who developed resistance, to support the development of new therapeutic strategies. Methods: Fresh tumor biopsy specimens were prospectively collected from patients through the MATCH-R clinical trial (NCT02517892) or the UNLOCK program. PDX models were established in NOD Scid Gamma (NSG) mice, which were then developed and characterized. These models underwent immunohistochemistry, comprehensive molecular profiling (including whole exome sequencing [WES] and bulk RNA sequencing [RNA-seq]), and pharmacological validation. Results: As of November 2024, 197 PDX models have been successfully generated from 519 biopsies, with a global take rate of 38%. These models are developed across several cohorts, including castration-resistant prostate cancer (22 PDX: 18 AR pathway inhibitors, 1 post-PARPi, and 3 Lu-PSMA), lung cancer (40 PDX with EGFR inhibitors, including 28 post-osimertinib, and 22 PDX with ALK inhibitors, including 2 post-brigatinib, 9 post-lorlatinib, and 6 post-alectinib). Additionally, 34 PDX models have been developed for the FGFR cohorts with FGFR2/FGFR3 alterations across various cancers, including cholangiocarcinomas, bladder carcinomas, gynecological carcinomas, and pancreatic cancers. We also developed 32 PDX models with KRAS mutations (G12C, n=23; G12D, n=9) from various tumors, including lung, pancreatic, gynecological, and colon cancers. Recently, we launched a new UNLOCK cohort focused on antibody-drug conjugate therapy, developing 12 additional PDX models. These PDX models faithfully replicate the genetic, transcriptional, phenotypic, and pharmacologic features of the original biopsies, providing a unique preclinical platform for testing novel drugs and combination therapies. Adaptive treatments with new or combinatorial strategies are being explored to restore drug sensitivity. Conclusion: The development of 197 PDX models under the UNLOCK program represents a significant resource, enabling the evaluation of novel drugs, adaptive treatment strategies, and combinatorial approaches. This unique collection of PDX models paves the way for advancing precision oncology and overcoming therapeutic resistance. Citation Format: Ludovic Bigot, Nobre Nobre, Alice Da silva, Melissandre Meteau, Mathis Delavigne, Miguel Soares, Francesco Facchinetti, Loic Poiraudeau, Floriane Braye, Inmaculada Inmaculada Alonso Garcia, Natalie Ngoi, Alexandre Halimi, Emlie Natali, Damien Vasseur, Kristi Beshiri, Claudio Claudio Nicotra, Maud Ngo-Camus, Lambros Tselikas, Beatriz Beatriz Alonso de Castro, Julieta Rodriguez, Maria Fernanda Mosele, Maud Kamal, Gérôme Jules-Clement, Karim Fizazi, Christophe Massard, Benjamin BESSE, Luc Friboulet, Yohann Loriot. UNLOCK, a preclinical platform of PDX models resistant to innovative therapies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1276.
Read full abstract