In metastatic cancer, the role of heterogeneity at the tumor-immune microenvironment, its molecular underpinnings and clinical relevance remain largely unexplored. To understand tumor-immune dynamics at baseline and upon chemotherapy treatment, we performed unbiased pathway and cell type-specific immunogenomics analysis of treatment-naive (38 5 samples from 8 patients) and paired chemotherapy treated (80 paired samples from 40 patients) high-grade serous ovarian cancer (HGSOC) samples. Whole transcriptome analysis and imagebased quantification of T cells from treatment-naive tumors revealed ubiquitous variability in immune signaling and distinct immune microenvironments co-existing within the same individuals and within tumor deposits at diagnosis. To systematically explore cell type 10 composition of the tumor microenvironment using bulk mRNA, we derived consensus immune and stromal cell gene signatures by intersecting state-of-the-art deconvolution methods, providing improved accuracy and sensitivity when compared to HGSOC immunostaining and leukocyte methylation data sets. Cell-type deconvolution and pathway analyses revealed that Myc and Wnt signaling associate with immune cell exclusion in untreated HGSOC. To evaluate 15 the effect of chemotherapy on the intrinsic tumor-immune heterogeneity, we compared sitematched and site-unmatched tumors before and after neoadjuvant chemotherapy.Transcriptomic and T-cell receptor sequencing analyses showed that site-matched samples had increased cytotoxic immune activation and oligoclonal expansion of T cells after chemotherapy, which was not seen in site-unmatched samples where heterogeneity could not be accounted 20for. These results demonstrate that the tumor-immune interface in advanced HGSOC is intrinsically heterogeneous, and thus requires site-specific analysis to reliably unmask the impact of therapy on the tumor-immune microenvironment. 1050 Conceptualization [Ideas; formulation or evolution of overarching research goals and aims] AS, AJS, MLM, ES Data curation [Management activities to annotate (produce metadata), scrub data and maintain research data (including software code, where necessary for interpreting the data itself) for 1055 initial use and later re-use] AJS, KL, PC Formal analysis [Application of statistical, mathematical, computational, or other formal techniques to analyse or synthesize study data] 1060 AJS Funding acquisition [Acquisition of the financial support for the project leading to this publication] AV, ES, AS, MLM 1065 Investigation [Conducting a research and investigation process, specifically performing the experiments, or data/evidence collection] PC, KL, TW, YM, IN, BW, DC, ES 1070 50 Methodology [Development or design of methodology; creation of models] ES Project administration [Management and coordination responsibility for the research activity planning and execution] 1075 AS, MLM, ES Resources [Provision of study materials, reagents, materials, patients, laboratory samples, animals, instrumentation, computing resources, or other a...