Abstract Bladder, colon, gastric, prostate, and uterine cancers originate from the organ’s epithelium, and each organ is surrounded by smooth muscle. In prostate cancer, organ-confined tumors, without extracapsular extension (ECE) through muscle, have a 5-year survival rate of 99% compared to 31% for metastatic disease. Previously, we modeled tumor cluster ECE and reported that a CRISPR-Cas9 integrin mutation (α6AA) blocked ECE by switching from a cell-ECM to a cell-cell biophysical phenotype. Since tumors are mosaics and migrate as heterogeneous drug-resistant populations, we tested the functional heterogeneity of the two biophysical cell phenotypes in ECE. Experimental procedures included the in vivo mouse xenograft model of ECE as a functional endpoint of muscle invasion, the MTT assay for cell survival, and using electric cell impedance sensing (ECIS) measurements to directly compare the biophysical properties of tumor cells expressing the invasion permissive α6WT integrin, cells expressing only the α6AA mutation, or a mixed population expressing both integrin types. The α6AA integrin mutation or α6KO knockout produced cells unable to invade into and through the smooth muscle of the mouse as compared to the α6WT integrin cells, as previously reported by us. The new unpublished results are that cell adhesion mediated drug resistance was detected in the α6AA population since their LD50 to Bortezomib, Gemcitabine, and Taxotere was shifted to 33.71nM, 64.66nM, and 5.77nM respectively from 16.29nM, 31.74nM, and 3.52nM in the α6WT cells as measured by MTT assay at 72 hours of incubation. In contrast, the α6AA cells were sensitized to the NAMPT inhibitor, FK866, with LD50 shifting from 27.46nM in α6WT to 7.354nM in the α6AA cells. A mixture of α6AA and α6WT cells allowed invasion into and through the muscle in the mouse of the mutant α6AA integrin cells that could not invade on their own. Examining the live cellular biophysical parameters revealed that the integrin mutation produced a 4-fold increase in cell-cell resistance (400Hz measurement) and a 12-fold recovery time delay in re-establishing a cell-cell resistance monolayer after wounding, with response complexity decreased two-fold. Cell-ECM resistance (40,000Hz measurement) in α6AA versus α6WT population was 1.5-fold decreased with a 6-fold recovery time delay in the cell-ECM resistance monolayer and response complexity decreased 2-fold. A heterogenous population containing up to a 4:1 proportion of α6AA to α6WT cells synergistically delayed cell-cell recovery time of the wounded monolayer up to 6-fold compared to the α6WT cells. Conversely, recovery of the wounded cell-ECM resistance monolayer with 1:1 or 4:1 mixture of α6AA cell only delayed recovery time by 1.5–1.7-fold. Taken together, these data suggest that a heterogenous tumor population provides functional advantages for drug-resistant tumor cells to invade and traverse the muscle barrier. Current work is ongoing to optimize the drug combinations to eliminate both phenotypes in the invasive tumor network and prevent ECE. Citation Format: Kendra D. Marr, Jaime M.C. Gard, William L. Harryman, Elijah J. Keeswood, Allan I. Paxson, Charles Wolgemuth, Lori A. Hazlehurst, Raymond B. Nagle, Anne E. Cress. Heterogeneity of cancer network biophysical phenotypes is required for tumor muscle invasion in vivo [abstract]. In: Proceedings of the AACR Special Conference: Advances in Prostate Cancer Research; 2023 Mar 15-18; Denver, Colorado. Philadelphia (PA): AACR; Cancer Res 2023;83(11 Suppl):Abstract nr B001.