Abstract Introduction: Drug candidates advance to phase I clinical trial after rigorous optimization at preclinical stage and yet nine out of ten drug candidates fail during clinical trials and approval stage. Lack of clinical efficacy is a major contributor (~ 50%) towards this 90% of the failure at clinical trials. Tremendous efforts are being taken to demonstrate drug efficacy in preclinical studies by mitigating the biological discrepancy of molecular target between human disease and the model used for the validation. Herein this study, we aim to showcase therapeutic responses of heat-shock protein 90 (HSP90) inhibitor regimens (HSP90i only and HSP90i + Doxorubicin combination) against three-dimensional (3D), patient-derived tumoroid (PDT) model platforms with progressing tumor microenvironment (TME) representability by incorporating supplementary TME structure, immunity, and oxygenation. Revelations on efficacy and resistance patterns of such drug treatments on increasingly complex models is aimed at providing unique insights to tumor and drug-drug interplay. Methods: Human Uterine Adenosarcoma PDT models were prepared across three COMPASS (Custom Organoid Modeling Platform for Accurate SolutionS) platforms: 1) COMPASS 1a - Non-scaffold 2) COMPASS 3b - Scaffold + Immune cells and 3) COMPASS 3d - Scaffold + Immune cells + Oxygenation. The various platforms were tested against two regimens A) HSP90i only [50 nM and 100 nM] and B) HSP90i [50 nM and 100 nM] + Doxorubicin [50 nM]. Tumor size and volume assessment were carried out on Days 3 and 7 post-treatment -with statistical done via one-way ANOVA, followed by student’s t test for significance (p<0.05, n=3). Results: While a simpler platform COMPASS 1a showcased the therapeutic efficacy of monotherapeutic HSP90i [at 100 nM] (p<0.05), progressively complex models (COMPASS 3b and COMPASS 3d) have otherwise indicated the lack of significant tumor volume changes. On the other hand, current findings on combination therapies of HSP90i and Doxorubicin have revealed persistent therapeutic resistance (even suggestive of patterns of increasing tumor growth) on COMPASS 1a, necessitating further comparisons on the more complex models. Conclusion: Our results demonstrate the need to test drugs in context of TME and oxygenation to help stratify patients to develop better clinical trials based on patient tumor characteristics of vascularization and immunogenicity. We plan to test several different drug modalities and cancer types for these parameters to demonstrate the need and effectiveness of testing prior to taking the drugs to the clinic. Citation Format: Huckie Del Mundo, Damieanus O. Ochola, Bhuvanesh Dave, Preethi Samuel. Patient-derived tumoroid platforms with progressing tumor microenvironment representability reveals efficacy and resistance patterns in pre-clinical HSP90i regimen testing for improved clinical insights [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 4230.