Abstract Background: EGFR tyrosine kinase inhibitors are effective for treatment of non-small cell lung cancer. Benefit from these therapies is often transient due to residual disease leading to acquired resistance. Osimertinib now used in first-line setting, but mechanisms of resistance are ill-defined and no effective second-line therapies exist. Disease progression occurs through tumor evolution that involves the emergence of distinct genetic alterations and non-genetic changes in cell state. In this study, we determine the molecular drivers of drug sensitivity and tumor evolution which forms the basis of clinical genomic heterogeneity. We show that the increase in genomic instability correlates with increased chromosomal alterations or rate of mutation which creates evolutionary dependencies. Method: Using our previously established EGFR-TKI acquired resistance models, we observed heightened TPX2/AURKA activity. We measured the temporal activation of TPX2/AURKA in sensitive phase, a drug-tolerant phase (residual phase) and acquired resistant phase. Chromosomal segregation errors, mitotic deformation and aneuploidy were used as a functional read-out to the measure genomic instability upon EGFR-TKI treatment. Using clinical specimens, the contribution of non-genetic TPX2/AURKA axis in disease progression was evaluated. Results: In-vitro models of residual disease exhibited increased AURKA activation via upregulation of its co-activator TPX2. Pharmacological inhibition of AURKA activation by MLN8237 together with osimertinib enhanced the magnitude of response and forestalled the emergence of resistance. Therefore, tumor cells reply on AURKA to transition from a sensitive to drug-tolerant phase and is maintained in acquired resistance. Since high levels of TPX2 and AURKA are known to cause mitotic errors and polyploidy, we surveyed for mitotic defects induced by EGFR-TKI treatment. EGFR-TKI resulted in a marked accumulation of errors in centrosome biogenesis, spindle assembly, and chromosome segregation resulting in polyploid cells with abnormal DNA content indicating persistent mitotic stress is a feature of EGFR inhibition. We conclude that the abnormal activation of TPX2/AURKA leaves a signature of defects associated with mitotic stress, a prominent feature of genomic instability. Examination of clinical samples with T790M mutation or MET amplification upon disease progression had high levels of TPX2. This nominates TPX2 as a novel biomarker therapeutic resistance in a significant fraction of EGFR-mutant lung cancers. Conclusion: Our results indicate that heightened TPX2/AURKA activation is required for tumor cells to transition out of sensitive cell state and persist upon drug pressure. This deregulation creates a highly genomically unstable environment and give rise to genomic heterogeneity providing a fertile ground for the subsequent survival of fittest subclones. Citation Format: Khyati N. Shah, Roma Bhatt, Julia Rotow, Julia Rohrberg, Victor Olivas, Victoria E. Wang, Jonathan Kuhn, Sophie Dumont, Frank Mccormick, Andrei Goga, Collin M. Blakely, Trever G. Bivona, Sourav Bandyopadhyay. Non-genetic TPX2/AURKA signaling facilitates tumor evolution in EGFR-TKI resistance in NSCLC [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2902.