Abstract 7130: Spatially resolved single-cell analysis reveals cGAS-STING-mediated tumor-immune interactions via IFNα as determinants of pembrolizumab sensitivity in triple-negative breast cancer
Bibliographic record
Abstract
Abstract Purpose: While an addition of immune checkpoint inhibitor (ICI), pembrolizumab, has improved outcomes of patients with TNBC, early-stage TNBC patients with residual cancer burden (RCB) II/III after neoadjuvant chemotherapy in combination with pembrolizumab (NAP) still have poor outcomes. Previously, we demonstrated that intratumoral HLA-DRA expression and interferon-alpha (IFNα) responses were associated with immune-inflamed phenotype and better outcomes in TNBC patients receiving chemotherapy. This study utilized the CosMx™ SMI to investigate in early-stage TNBC treated with NAP. Methods: Of 151 samples from TNBC patients treated with NAP at Mayo Clinic, 8 pre-treatment biopsy samples (5 pCR and 3 RCB II/III were analyzed using the CosMx™ SMI platform. Linear mixed models to assess differential expression (DE), adjusting for hierarchical data structures. DE analyses were expressed as log2-fold changes (logFC), with the Benjamini-Yekutieli method. GSEA and IPA for signaling pathway exploration. Results: 1.7 million cells were identified across 8 samples after stringent quality control. Immune cell profiling revealed higher B cells, plasma cells, plasmablasts, and T cells (CD4 and CD8) in responders, whereas fibroblasts and myeloid cells were enriched in non-responders. DE analysis showed significantly elevated expression of HLA-DRA (logFC 1.955, p<0.001), other MHC class I and II components, and IFNα in responders. Tumor cells in responders exhibited the highest IFNα expression across cell types. GSEA and IPA analysis revealed enrichment in interferon signaling, antigen presentation, and the cGAS-STING pathway, which drives IFNα production. Spatial analysis categorized cells by their proximity to tumor cells: within 50 µm, 50-100 µm, and over 100 µm. T cells, particularly CD8 T cells, showed a distance-dependent distribution, with a higher abundance near tumor cells in responders. CD8 T cell activation scores declined with increasing distance from IFNα-positive tumor cells but remained consistently low near IFNα-negative tumor cells. IFNγ-positive CD8 T cells exhibited higher activation compared to IFNγ-negative cells, with activation levels influenced by proximity to tumor cells. Lastly, MHC expression in tumor cells was examined relative to CD8 T cell proximity. HLA-DRA and CIITA expression decreased as tumor cells were farther from CD8 T cells. Conclusions: This spatially resolved single-cell analysis highlights the pivotal role of tumor-immune interactions in response to pembrolizumab in TNBC. Our data suggest that tumor cells secrete IFNα through the activation of the cGAS-STING pathway, which activates T cells within close proximity, highlighting the potential role of cGAS-STING agonist as well as IFNα signaling and antigen presentation in augmenting TNBC responses to ICIs. Citation Format: Yi Liu, Saranya Chumsri, Yaohua Ma, Jodi M. Carter, Aziza Nassar, Edith Perez, Roberto A. Leon-Ferre, David Zahrieh, David W. Hillman, Judy C. Boughey, James Ingle, Krishna R. Kalari, Fergus J. Couch, Matthew P. Goetz, Keith L. Knutson, E. Aubrey Thompson. Spatially resolved single-cell analysis reveals cGAS-STING-mediated tumor-immune interactions via IFNα as determinants of pembrolizumab sensitivity in triple-negative breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7130.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".