Abstract 463: Integrated digital pathology and transcriptome analysis identifies molecular mediators of T cell exclusion in ovarian cancer
Bibliographic record
Abstract
Background:Close proximity between cytotoxic T lymphocytes and tumor cells is required for effective immunotherapy. Three tumor-immune (TI) phenotypes, infiltrated, excluded and desert, have been previously described based on the infiltration patterns of CD8+ T cells. However, no quantitative methods exist to define these phenotypes robustly in human solid tumors. Importantly, the molecular features and mechanisms determining these phenotypes are not well understood. Here we report a novel integrated approach to classify and functionally dissect TI phenotypes in human ovarian cancer.Methods:CD8 IHC and RNAseq analysis were performed on 370 ovarian tumors from the ICON7 phase III clinical trial, a front-line trial testing the addition of bevacizumab to chemotherapies. A digital image analysis algorithm was developed to quantify the quantity and spatial distribution of CD8+ T cells. Coupling digital pathology with transcriptome analysis, a random forest machine learning algorithm was applied to identify genes associated with these two metrics using a training set (n=155). A gene expression-based classifier was developed for classifying TI phenotypes and validated using testing sets from ICON7 trial and a vendor collection. Functional characterization of key mediators promoting T cell exclusion were carried out by integrating in situ, in vitro and ex vivo analyses on ovarian tumor tissues, cancer associated fibroblasts (CAFs) and ovarian cancer cell lines. Anti-tumor activity of TGFβ blockade in combination with anti-PD-L1 was evaluated in the mouse BrKras ovarian cancer model in FVB background.Results:Integrating digital pathology and machine learning on large ovarian tumor cohorts, we developed and validated a 157-gene molecular classifier. We show the TI phenotypes are of biological and clinical importance in ovarian cancer. Two hallmarks of T cell exclusion were identified: 1) loss of MHC I on tumor cells and 2) upregulation of TGFβ/stromal activities. We show that MHC I in ovarian cancer cells is likely regulated by epigenetic mechanisms and TGFβ is a key mediator of T cell exclusion. TGFβ reduced MHC I expression in ovarian cancer cells and induced extracellular matrix and immunosuppressive molecules in human primary fibroblasts. Finally, we demonstrated that combining anti-TGFβ and anti-PD-L1 in the BrKras mouse model improved the anti-tumor efficacy and survival.Conclusion:This study provided the first systematic and in-depth characterization of the molecular features and mechanisms underlying the tumor-immune phenotypes in human ovarian cancer. We illuminated a multi-faceted role of TGFβ in mediating crosstalk between tumor cells and CAFs to shape the tumor-immune contexture. Our findings support that targeting the TGFβ pathway represents a promising therapeutic strategy to overcome T cell exclusion and optimize response to cancer immunotherapy.Citation Format: Melanie Desbois, Akshata Udyavar, Lisa Ryner, Cleopatra Kozlowski, Yinghui Guan, Milena Dürrbaum, Shan Lu, Jean-Philippe Fortin, Hartmut Koeppen, James Ziai, Ching-Wei Chang, Amy Lo, Shilpa Keerthivasan, Marie Plante, Richard Bourgon, Carlos Bais, Priti Hegde, Anneleen Daemen, Shannon Turley, Yulei Wang. Integrated digital pathology and transcriptome analysis identifies molecular mediators of T cell exclusion in ovarian cancer [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 463.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.051 | 0.030 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; both teacher heads agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".