Decoding the Heterogeneity in Chemo-Immunomodulation to Unlock the Potential of Chemoimmunotherapy in Metastatic Triple-Negative Breast Cancer

PROJECT SUMMARY/ABSTRACT Triple-negative breast cancer (TNBC) is a highly heterogeneous and aggressive subtype of breast cancer characterized by the highest potential for metastasis and thus the worst clinical outcomes of all breast cancer subtypes. The outcome in metastatic TNBC (mTNBC) is particularly poor with a median overall survival time of < 16 months. Chemoimmunotherapeutic treatment strategies employing chemotherapeutics, including both conventional compounds and targeted therapies, and PD-1 blockade have proven efficacious for a variety of solid tumor malignancies, however, the efficacy of anti-PD-1 chemoimmunotherapy is limited in the context of mTNBC with an estimated overall progression-free survival of < 10 months and overall survival < 2 years for the best-case scenarios. While the outcomes observed thus far represent a minor improvement for mTNBC, anti- PD-1 chemoimmunotherapy has achieved > 40% 5-year overall survival with greater improvements in response rate and progression-free survival in other solid tumor malignancies. Thus, further studies are needed to optimize anti-PD-1 chemoimmunotherapeutic strategies in mTNBC. Despite the essential synergistic role that chemotherapy plays in anti-PD-1 chemoimmunotherapeutic treatment strategies, the identification and selection of the optimal chemotherapeutic agent(s) that best synergize with PD-1 blockade for an individual patient remain a critically unmet need. Previous studies have identified that the immunomodulatory effects of chemotherapy (chemo-immunomodulation) are imperative for anti-PD-1 chemoimmunotherapeutic efficacy, thus, understanding the factors that influence chemo-immunomodulation may be critical for optimizing anti-PD-1 chemoimmunotherapeutic treatment strategies. Nevertheless, while studies have sought to understand genomic and transcriptomic features that influence chemoresistance, few studies have focused on factors that influence chemo-immunomodulation. The objectives of this proposal are to identify, characterize, and establish the clinical relevance of genomic and transcriptomic features that influence chemo-immunomodulation and utilize the data generated from the studies herein to develop a machine-learning- based model predictive of chemo-immunomodulation by select agents. Aim 1, 2A, and 2B of this proposal will identify and characterize genomic and transcriptomic factors that influence chemo-immunomodulation using bulk and single-cell RNA-sequencing approaches in in vitro, in vivo, and ex vivo models of mTNBC. Aim 2C will establish the clinical relevance of these findings for selected chemotherapeutics using preclinical mTNBC murine models. Aim 3 seeks to utilize the abundance of data generated through single-cell RNA-sequencing to generate preliminary machine learning models that are predictive of chemo-immunomodulation for selected chemotherapeutic agents. The results of this study may provide information that guides novel approaches in designing, optimizing, and maximizing the efficacy of anti-PD-1 chemoimmunotherapeutic strategies based on patient-specific genomic and transcriptomic characteristics and their influence on chemo-immunomodulation.